Search Results

Search found 9861 results on 395 pages for 'embedded systems'.

Page 361/395 | < Previous Page | 357 358 359 360 361 362 363 364 365 366 367 368  | Next Page >

• How to Setup Ubuntu Mail Server with Google Apps?

  - by Apreche

  I have a domain, let's call it foobar.com. All of the MX records for foobar.com point to Google's mail servers because I am using Google Apps for your domain to manage it. It's great because everyone gets all the advantages of GMail, but our e-mail addresses aren't @gmail.com. I also have a server. Primarily, it's a web server, but it also serves other things. One of the things it serves is the web site for foobar.com and also sites for various virtual hosts such as shop.foobar.com and forum.foobar.com. The server is running Ubuntu 8.04, because I like using LTS releases in production. The thing is, there are various applications running on the server that need the ability to send out emails. Various applications, like the cron jobs, send me e-mails in case of errors. Some of the web applications need to send e-mail to users when they forget their passwords, to confirm new registered users, etc. Lastly, it's nice to be able to send e-mail from the command line using the mail command, or mutt. How can I setup the mail on the web server to go through the Google apps mail servers? I don't need the web server to receive mail, though that would be cool. I do need it to be able to send mail as any legitimate address @foobar.com. That way the forum application can send mails with [email protected] in the from field, and the ecommerce application will have [email protected] in the from field. Also, by sending the mail through the Google servers, we can avoid a lot of the problems with the e-mails being blocked by various spam filters on the web. Google's SMTP servers are trusted a lot more than mine would be. I'm pretty good with administering Linux systems, but I am absolutely brain dead when it comes to e-mail. I need step by step directions from beginning to end on how to set this up. I need to know every thing to install, and every single change to the configuration files that is necessary. I have tried following various howtos and guides in the past, but none of them were quite right. Either they didn't work at all, or they offered a configuration that is not what I wanted. Please help. Thanks.

  Read the article

• Linux Software RAID1 Rebuild Completes, but after reboot, its degraded again

  - by zimmy6996

  I have been beating my head with an issue here, and I'm now turning to the internet for help. I have a system running Mandrake Linux, with the following configuration: /dev/hda - This is a IDE drive. Has some partitions on it that boot the system and make up most of the file system. /dev/sda - This is drive 1 of 2 for a software raid /dev/md0 /dev/sdb - This is drive 2 of 2 for a software raid /dev/md0 md0 gets mounted but fstab as /data-storage, so it is not critical to the systems ability to boot. We can comment it out of fstab, and the system works just fine either way. The problem is, we have a failed sdb drive. So I shut the box down, and have pulled the failed disk and installed a new disk. When the system boots up, /proc/mdstat shows only sda as part of the raid. I then run the various command to rebuild the RAID to /dev/sdb. Everything rebuilds correctly, and upon completion, you look at /proc/mdstat and it shows 2 drives sda1(0) and sdb1(1). Everything looks great. Then you reboot the box ... UGH!!! Once rebooted, sdb is missing again from the RAID. It is like the rebuild never happened. I can walk through the commands to rebuild it again, and it will work, but again, after reboot, the box seems to make sdb just vanish! The real odd thing is, if after reboot, I pull sda out of the box, and try to get the system to load with the rebuilt sdb drive in the system, and when I do, the system actually throws and error just after grub, and says something about drive error, and the system has to shut down. Thoughts??? I'm starting to wonder if grub has something to do with this mess. That the drive isn't being setup within grub to be visible at boot? This RAID array isn't necessary for the system to boot, but when the replacement drive is in there, without SDA it won't boot system, so it makes me believe there is something to that. On top of that, there just seems to be something wonky here the drive falling off of RAID after reboot. I've hit the point of pounding my head on the keyboard. Any help would be greatly appreciated!!!

  Read the article

• File/printer sharing issues on network with multiple OSes

  - by DanZ

  My workplace consists of computers running a variety of different operating systems, and I have been running into problems getting some of them to connect to a shared drive and printer over the network. Here is a brief description of the computers involved and the issues I have encountered: 1: Dell desktop, Windows Vista Business-- This is the computer I want the others to connect to. It has a USB printer and eSATA hard drive enclosure that I have set up for sharing, with different accounts for the various users. 2: Fujitsu laptop, Windows XP Tablet edition-- No problems. Can connect to both the shared printer and hard drive. 3: Lenovo laptop, Windows Vista Business 64 bit-- No problems. Can connect to both the shared printer and drive. 4: Apple MacBook, OS 10.4-- Can connect to the shared drive, but not to the shared printer. I am aware that the printer issue is due to a known incompatibility between Vista and OS 10.4 and earlier with regards to Samba. It is not a big problem, however, as this computer can access a network printer. 5: Sony laptop, Windows Vista Home Premium-- Can connect to the shared printer, but not the shared drive. It can see computer 1 and its shared drive on the network, and appears to successfully log in to user accounts. However, if you try to access the shared drive, it says you do not have permission. I have tried both standard and administrator accounts, and none can access the drive from this computer. 6: MacBook Pro, OS 10.5 (there are two of these)-- Can connect to the shared printer, but not the shared drive. They can't see computer 1 on the network. For that matter, they also can't see each other or the older Mac, but can see and access shared folders on the XP machine (computer 2) and can see other PCs in the building. I was able to add the shared printer manually by typing in its network location, but was unable to manually add the shared drive in the same way. So, what I am looking for is suggestions on how to get computers 5 and 6 to connect to the shared drive. Since they can already connect to the shared printer (which is on the same computer as the shared drive), it seems reasonable that they should be able to access the drive as well.

  Read the article

• How to get the best LINPACK result and conquer the Top500?

  - by knweiss

  Given a large Linux HPC cluster with hundreds/thousands of nodes. What are your best practices to get the best possible LINPACK benchmark (HPL) result to submit for the Top500 supercomputer list? To give you an idea what kind of answers I would appreciate here are some sub-questions (with links): How to you tune the parameters (N, NB, P, Q, memory-alignment, etc) for the HPL.dat file (without spending too much time trying each possible permutation - esp with large problem sizes N)? Are there any Top500 submission rules to be aware of? What is allowed, what isn't? Which MPI product, which version? Does it make a difference? Any special host order in your MPI machine file? Do you use CPU pinning? How to you configure your interconnect? Which interconnect? Which BLAS package do you use for which CPU model? (Intel MKL, AMD ACML, GotoBLAS2, etc.) How do you prepare for the big run (on all nodes)? Start with small runs on a subset of nodes and then scale up? Is it really necessary to run LINPACK with a big run on all of the nodes (or is extrapolation allowed)? How do you optimize for the latest Intel/AMD CPUs? Hyperthreading? NUMA? Is it worth it to recompile the software stack or do you use precompiled binaries? Which settings? Which compiler optimizations, which compiler? (What about profile-based compilation?) How to get the best result given only a limited amount of time to do the benchmark run? (You can block a huge cluster forever) How do you prepare the individual nodes (stopping system daemons, freeing memory, etc)? How do you deal with hardware faults (ruining a huge run)? Are there any must-read documents or websites about this topic? E.g. I would love to hear about some background stories of some of the current Top500 systems and how they did their LINPACK benchmark. I deliberately don't want to mention concrete hardware details or discuss hardware recommendations because I don't want to limit the answers. However, feel free to mention hints e.g. for specific CPU models.

  Read the article

• Searching For a Desktop Security Software to harden Windows machines, anybody?

  - by MosheH

  I'm a network administrator of a small/medium network. I'm looking for a software (Free or Not) which can harden Windows Computers (XP And Win7) for the propose of hardening standalone desktop computers (not in domain network). Note: The computers are completely isolated (standalone), so i can't use active directory group policy. moreover, there are too many restriction that i need to apply, so it is not particle to set it up manual (one by one). Basically what I’m looking for is a software that can restrict and disable access for specific user accounts on the system. For Example: User john can only open one application and nothing else -- He don’t see no icon on the desktop or start menu, except for one or two applications which i want to allow. He can't Right click on the desktop, the task-bar icons are not shown, there is no folder options, etc... User marry can open a specific application and copy data to one folder on D drive. User Dan, have access to all drives but cannot install software, and so on... So far ,I've found only the following solutions, but they all seems to miss one or more feature: Desktop restriction Software 1. Faronics WINSelect The application seems to answer most of our needs except one feature which is very important to us but seems to be missing from WINSelect, which is "restriction per profile". WINSelect only allow to set up restrictions which are applied system-wide. If I have multiple user accounts on the system and want to apply different restrictions for each user, I cant. Deskman (No Restriction per user)- Same thing, no restriction per profile. Desktop Security Rx - not relevant, No Win7 Support. The only software that I've found which is offering a restriction per profile is " 1st Security Agent ". but its GUI is very complicated and not very intuitive. It's worth to mention that I'm not looking for "Internet Kiosk software" although they share some features with the one I need. All I need is a software (like http://www.faronics.com/standard/winselect/) that is offering a way to restrict Windows user interface. So if anybody know an Hardening software which allows to set-up user restrictions on Windows systems, It will be a big, big, big help for me! Thanks to you all

  Read the article

• open source knowledge base CMS system

  - by Thomi

  I'm looking for an open source knowledge base system that uses tags, rather than free-text search to identify articles (a lot like serverfault does). I've looked at twiki, which many people suggested, but haven't found what I'm looking for. Basically I want to be able to create and tag articles, and provide an easy way for anonymous users to search based on tags. Edit: OK, here's some more detail regarding what I want. Basically, all the knowledge base systems I have seen so far are a collection of articles, each article with a title. Most of them allow you to categorise articles into groups and sub-groups. Users of the system can search for information using a title search, for example "How do I print from AwesomeProduct?" - which then shows a list of any articles that match that search text. This is fine and dandy when your KB is for one version of the software product (the mythical AwesomeProduct ver 1.0). However, the development team then go ahead and create a new version (ver 2.0) that adds many new features and changes some existing features. Now, how do we support both products in the same KB? The Naive method is to copy all articles from 1.0, and update them for 2.0, adding and removing articles in 2.0 as required. We can then add text at the top of every 1.0 article that says: "this articles applies to 1.0 only, to see the 2.0 version, click here" (or something similar) The problem with articles being indexed in the system by title is that it's very hard to filter based on meta-data like version. What happens when we create version 3.0 or 4.0? The end-situation here is that you have a mess of articles. They're hard to search, hard to filter, and even harder to manage. The solution (it seems to me) is to use tags, rather than text as the article index mechanism. So articles can be tagged with a tag representing the software version, topic area etc. etc. Users can then filter based on tag - an example search might be "version_1 printing" - which straight away gives a list of articles with all these tags. So that's what I'm looking for - a KB system that uses tags, rather than text to index many articles. I'm sure I could build something with drupal, but I was hoping for something that worked out-of-the-box.

  Read the article

• Random Computer Crashes

  - by Josh W.

  Ok, here's a wierd one for you all. Occasionally my PC here at work will crash in a very peculiar way. My dual monitors will suddenly go blank as if there is no longer a video signal, the USB mouse light will go dark and mouse stays unresponsive, the keyboard lights will not change status when the appropriate keys are pressed (Num/Caps/Scoll Lock). The CD Tray WILL open and close. But the computer will not respond to a ping request. For all intents & purposes it's as if the computer is off, except it wasn't intentional by me. The power light and internal fans are still on and I've now lost any unsaved work. Now here's where it gets wierd. This PC is part of a batch of PC's we got from a local vendor who does our initial system builds. Mine, and 6 other co-workers PCs all have the same issue. Originally we thought it was a bad combination of hardware, but through trial and error the only thing we haven't eliminated are the OS, Mobo & CPU. The problem was so bad for some of them that they ended up going back to their 5 year old dinosaurs in order to get some work done, for me the problem isn't as bad, maybe once every other day or so, but still enough to bite me in the ass if I've forgotten my ritualistic pressing of CTRL-S every 1-5 minutes. In this case we've tried two different video cards, two different power supplies, two different memory configurations, running on a UPS/not on UPS, updating/rolling back video drivers, three different bios revisions. The only things we haven't swapped are the mobo & cpu, mainly because a new mobo means a new Hardware Abstraction Layer, ie re-install of windows and there's alot of other software on this PC that takes forever to reload by hand. There was a base image that our systems team created with all the drivers installed and the basic setup of software our company uses, but they then must customize the setup for us programmers so it takes a while to get a new configuration up and going. I'm a programmer by day and am usually pretty good at diagnosing computer problems whether through trial and error or not. We've pretty much exhausted all the ideas we can think of here, short of a new mobo/cpu. Was hoping someone out there might have anything else we can try.. Relevant Parts: OS: XP Pro 32-bit Motherboard: Intel DG41RQ CPU: Intel Core-2 Quad Q9400 @ 2.66GHz Current BIOS Version/Date Intel Corp. RQG4110H.86A.0014.2010.0306.1151, 3/6/2010 Dual LCD's, Viewsonic VG930m & Samsung SyncMaster 910v (other people have different models, but listed in case there's some very wierd problem with the signals being sent/received) PS/2 Keyboard USB Microsoft Intellimouse BIOS Versions Tried: R 0013 12/23/2009 R 0014 3/6/2010 Video cards Tried GeForce 8400 GS Radeon HD 4350 - ASUS EAH4350 Two Different Power Supplies a 380W & 550W Ram Configurations 2GB - 1 x 2GB 4GB - 2 x 2GB

  Read the article

• MongoDB and datasets that don't fit in RAM no matter how hard you shove

  - by sysadmin1138

  This is very system dependent, but chances are near certain we'll scale past some arbitrary cliff and get into Real Trouble. I'm curious what kind of rules-of-thumb exist for a good RAM to Disk-space ratio. We're planning our next round of systems, and need to make some choices regarding RAM, SSDs, and how much of each the new nodes will get. But now for some performance details! During normal workflow of a single project-run, MongoDB is hit with a very high percentage of writes (70-80%). Once the second stage of the processing pipeline hits, it's extremely high read as it needs to deduplicate records identified in the first half of processing. This is the workflow for which "keep your working set in RAM" is made for, and we're designing around that assumption. The entire dataset is continually hit with random queries from end-user derived sources; though the frequency is irregular, the size is usually pretty small (groups of 10 documents). Since this is user-facing, the replies need to be under the "bored-now" threshold of 3 seconds. This access pattern is much less likely to be in cache, so will be very likely to incur disk hits. A secondary processing workflow is high read of previous processing runs that may be days, weeks, or even months old, and is run infrequently but still needs to be zippy. Up to 100% of the documents in the previous processing run will be accessed. No amount of cache-warming can help with this, I suspect. Finished document sizes vary widely, but the median size is about 8K. The high-read portion of the normal project processing strongly suggests the use of Replicas to help distribute the Read traffic. I have read elsewhere that a 1:10 RAM-GB to HD-GB is a good rule-of-thumb for slow disks, As we are seriously considering using much faster SSDs, I'd like to know if there is a similar rule of thumb for fast disks. I know we're using Mongo in a way where cache-everything really isn't going to fly, which is why I'm looking at ways to engineer a system that can survive such usage. The entire dataset will likely be most of a TB within half a year and keep growing.

  Read the article

• Issue using a "used" SSD as a Windows 8.1 Boot Drive

  - by EpiGrad

  So, I'm something of a Mac person, but decided to take a stab at this whole "build yourself a PC" thing - right now, the thing is assembled, posts just fine, and can get to the BIOS. The problem is the drive I want to use - I intended to use a 80 GB Corsair SSD I've had sitting around as the boot drive, and a new Samsung SSD for games and the like. So I boot using a Windows 8.1 install USB stick, and if the Samsung drive is plugged in, it happily offers to install Windows on it. The Corsair drive though, it's flipped out - I reformatted it as a blank NTFS drive (it was HFS for Mac purposes) and the BIOS can't see it, nor can the Windows installer. What's wrong, and how do I fix it? The tools at my disposal are: The current ASUS BIOS that came with my motherboard (a Z87I-Deluxe), a Mac running the latest OS X which can also boot to Windows 7 if needed via either Parallels or Bootcamp. Update 1: Update: Based on a friend's suggestion to switch SATA ports, Windows 8.1's installer can now see the drive as Drive 0, Partition 1, a 83.8 GB "Primary" partition. But when I click it and hit "Next", I get the following error: "We couldn't create a new partition or locate an existing one. For more information, see the Setup log files" - not that it gives any clue how to access those. Update 2: Following a trail of Google suggestions, I ended up going into advanced tools and just reformatting the drive as follows: Start DISKPART. Type LIST DISK and identify your SSD disk number (from 0 to n disks). Type SELECT DISK <n> where <n> is your SSD disk number. Type CLEAN Type CREATE PARTITION PRIMARY Type ACTIVE Type FORMAT FS=NTFS QUICK Type ASSIGN Type EXIT twice (one to get out of DiskPart, the other to exit the command line tool) Per these instructions. This goes well enough, but now I can select the disk for installation, and I get a new error: "Windows 8 cannot be installed to this disk. The selected disk has an MBR partition table. On EFI systems, Windows can only be installed to GP disks." So, Googling that, I do the following: select disk 0 clean convert gpt exit ...and we might have fixed it. Windows is at least trying to install now.

  Read the article

• SQL Transactional Replication snapshot not applying

  - by dmch2

  Hi, I'm using SQL Transactional Replication with pull subscriptions to replicate databases (hosting their own distribution database) from several servers across a VPN to a central server. I've got the first 2 databases working fine but the 3rd one is causing me problems. My subscription server is SQL 2008, the source systems are all SQL 2005. The source databases are a few 100Mb in size and contain audit data so are simply growing slowly by adding new records at approx 1kb a second. As far as the replication monitor, Agent logs and event logs show everything is working fine - except that no data appears in my subscription database. The distribution agent doesn't seem to want to read the snapshot (and hence the initial state and schema) from the publisher. New transactions aren't applied although they do seem to be arriving OK as the replication monitor shows things like '5 transactions with 10 commands were delivered'. I would expect (as in previous times) to see statements about data being BCPed in the replication monitor. The snapshot is on the publisher on a shared folder. The subscriber can view the snapshot OK (\\repldata) and the alt snapshot folder is pointing at it. But the distribution agent doesn't seem to be making an attempt to do read it. I tried changing the snapshot path to something that's incorrect and didn't even get an error saying that it couldn't access it. After lots of googling etc I found that sp_MSget_repl_commands is called by the subscriber on the distribution database on the publisher. Running a profiler I can see that it's only called for one agent Id. After a reinit it's called for sequence number 0x0 as expected so I thought that would mean it's would look for the snapshot. However, looking on the publisher I see that there's data for two agents - the snapshot agent and the log reader agent (which is being queries). So I guess I need to tell the distribution agent to get the data for both. But how? and more importantly - why? It worked fine on the other two servers I've replicated. I'm not an SQL novice but this is pretty much my first go at replication so don't be afraid to accuse me of missing something obvious/stupid! I can get log files (eg from the distribution agent) if you want but they don't seem to have any errors in them - it just starts up and starts applying log reader agent changes. Cheers Dave

  Read the article

• How to make Nginx fire 504 immediately is server is not available?

  - by Georgiy Ivankin

  I have Nginx set up as a load balancer with cookie-based stickiness. The logic is: If the cookie is NOT there, use round-robbing to choose a server from cluster. If the cookie is there, go to the server that is associated with the cookie value. Server is then responsible for setting the cookie. What I want to add is this: If the cookie is there, but server is down, fallback to round-robbing step to choose next available server. So actually I have load balancing and want to add failover support on top of it. I have managed to do that with the help of error_page directive, but it doesn't work as I expected it to. The problem: 504 (and the fallback associated with it) fires only after 30s timeout even if the server is not physically available. So what I want Nginx to do is fire a 504 (or any other error, doesn't matter) immediately (I suppose this means: when TCP connection fails). This is the behavior we can see in browsers: if we go directly to server when it is down, browser immediately tells us that it can't connect. Moreover, Nginx seems to be doing this for 502 error: if I intentionally misconfigure my servers, Nginx fires 502 immediately. Configuration (stripped down to basics): http { upstream my_cluster { server 192.168.73.210:1337; server 192.168.73.210:1338; } map $cookie_myCookie $http_sticky_backend { default 0; value1 192.168.73.210:1337; value2 192.168.73.210:1338; } server { listen 8080; location @fallback { proxy_pass http://my_cluster; } location / { error_page 504 = @fallback; # Create a map of choices # see https://gist.github.com/jrom/1760790 set $test HTTP; if ($http_sticky_backend) { set $test "${test}-STICKY"; } if ($test = HTTP-STICKY) { proxy_pass http://$http_sticky_backend$uri?$args; break; } if ($test = HTTP) { proxy_pass http://my_cluster; break; } return 500 "Misconfiguration"; } } } Disclaimer: I am pretty far from systems administration of any kind, so there may be some basics that I miss here. EDIT: I'm interested in solution with standard free version of Nginx, not Nginx Plus. Thanks.

  Read the article

• Achieving A 1600 x 900 Resolution For (Guest OS) Ubuntu Under (Host OS) Windows 7

  - by panamack

  I am running Ubuntu 11.10 as a guest OS using VirtualBox 4.1.16 installed on Windows 7 Ultimate. On my laptop I'd like to be able to run Ubuntu in full screen mode at 1600 x 900. I only have options within the virtual machine to select 4:3 display settings such as 1600 x 1200, 1440 x 1050 etc. I have guest additions installed. At the windows command prompt, I tried typing: VBoxManage setextradata "Virtual Ubuntu Coursera ESSAAS" "CustomVideoMode1" "1600x900x16" This didn't work, still no 1600 x 900 res available in Ubuntu. I tried this having read the following section of the VirtualBox help (this also says something about a 'video mode hint feature' not sure what this means): 9.7. Advanced display configuration 9.7.1. Custom VESA resolutions Apart from the standard VESA resolutions, the VirtualBox VESA BIOS allows you to add up to 16 custom video modes which will be reported to the guest operating system. When using Windows guests with the VirtualBox Guest Additions, a custom graphics driver will be used instead of the fallback VESA solution so this information does not apply. Additional video modes can be configured for each VM using the extra data facility. The extra data key is called CustomVideoMode with x being a number from 1 to 16. Please note that modes will be read from 1 until either the following number is not defined or 16 is reached. The following example adds a video mode that corresponds to the native display resolution of many notebook computers: VBoxManage setextradata "VM name" "CustomVideoMode1" "1400x1050x16" The VESA mode IDs for custom video modes start at 0x160. In order to use the above defined custom video mode, the following command line has be supplied to Linux: vga = 0x200 | 0x160 vga = 864 For guest operating systems with VirtualBox Guest Additions, a custom video mode can be set using the video mode hint feature. UPDATE 02.06.12 I've just tried creating a new virtual machine using the same original disk image I had been given. This had Guest Additions v 4.1.6 installed and provided me with the 1600 x 900 full screen display I want. It's after I then install Guest Additions v 4.1.16 (the version included with my VirtualBox installation) that my only choices are 4:3 displays e.g. 1600 x 1200. Seems this is the cause.

  Read the article

• Should I choose KVM/XEN over OpenVZ or use them together?

  - by Krystian

  I've got a dual xeon e5504 server, with [for now] only 8GB of ram. Storage is'n impressive either: 3x 146GB sas in raid5 + 500GB sata drives. Currently it works as a development server, but it's over speced for our needs and since our development methods changed through last 2 years we decided it will work as a production system for some of our applications + we would like to have a separate system for testing/research. Our apps are mainly web apps deployed on tomcats [plural as some of the apps require older versions] and connected to Postgres. I would like to have a production system, where only httpd+tomcat+db are setup and nothing else runs there. Sterile system. Apart from that, I would like a test system, where I can play with different JVM settings, deploy my test apps, play with tomcat/httpd settings and restart them without interfering with the production system. Apart from that, I would like to be able to play with different linux flavors, with newer kernels to test how they work etc. I know, this is not possible with OpenVZ and I would have to choose KVM for that. I am thinking about merging the two, and setting up a KVM to be able to work with different systems [linux only to be frank] + use openVZ to setup separate machines for my development needs. I would simply go with that, but reading here and there about the performance impact full virtualization has over containers and looking at the specs of my server makes me think twice about it. I don't want to loose too much performance, especially because of the nature of my apps [few JVMs running at the same time]. It will be my first time with virtualization, apart from using desktop virtualbox/vmserver. Although I am a fast learner I don't want to mess with the main system so much that it will break the production apps or make them crawl. Although they are more or less internal apps and they don't produce much load, they need to be stable. I've read, that KVM host is a normal linux installation and it allows to run normal processes on it. If that is so, does it allow to run openVZ as well? I mean... can I have KVM and OpenVZ running on the same system/kernel? Or do I have to setup another system to run OpenVZ containers? How much performance impact can this have for me? Will my hardware suffice? oh and one more thing... unfortunately I'm quite limited with the funds... I'm looking for a free solution only :/

  Read the article

• Wrapping a point-to-point link

  - by user3712955

  I'm using a pair of IP radios (non-WiFi) to bridge my office engineering LAN (172.0.0.0/8) to a lab in another building. The radios work fine, but they expose a web management interface I'd like to hide, and they also generate traffic (ARP, STP, and more) that I need to keep off my (very, very clean) LAN segments. I have some ARM-Linux boards (similar to Beagle/Panda/RasPi) running Ubuntu, and I've put one at each end of the link, between the radio and the LAN. Each of the boards has 2 wired Ethernet interfaces, eth0 and eth1. The LAN segments are connected to eth0, and the radios are connected to eth1. I'd like to accomplish the following: Keep radio-originated traffic off my LAN segments! Hide all services provided by the radio (web, ssh, etc.) Transparently pass all traffic between the LAN segments (including things like ARP). The above also applies to the ARM-Linux boards: No stray traffic my LAN from them either! I'd like the system to look like a switch: LAN packets arriving at one eth0 appear at the other. And neither eth0 should have an IP address: The working system should behave like a CAT6 cable with some latency (instead of ARM boards and radios). Unfortunately, I'm confused about how to properly configure the ARM Ubuntu systems. What I'm guessing I need is a bridge on each board (br0?) and a VLAN (vlan0 or eth0.0?) to isolate the LAN traffic from everything else as it passes through the ARM boards and the radios. Then I need some kind of a firewall to block sending anything out eth0 that isn't from the other eth0 (via the VLAN). I've looked at the ip and ebtables commands (especially -t broute). While the concepts sorta-kinda make sense, I'm completely lost in the details. Edit: In the perverse case that a system on one of my LAN segments has the same IP address as one of the radios, or as eth1 on the ARM-Ubuntu boards, a VLAN won't work. Which I believe means I need to tunnel all traffic between the two eth0 interfaces to get that "like a wire" behavior. Help? Finally, I'd like to have a way to temporarily expose services on the ARM boards (ssh) and the radios (web) for maintenance purposes. Ideally, it would expose an IP address with ssh available on port 22. Once connected, I figure I'd start an X11 session and run a browser on the ARM board to access the radios. Or something. I could login via the console to enable/disable this, or perhaps could use a GPIO to trigger a script. I feel I've identified most of the pieces needed to make all this happen, but I have no idea how to combine them to make a working system. Thanks!

  Read the article

• FTP script needs blank line

  - by Ones and Zeroes

  I am trying to determine the reason for some FTP servers requiring a blank line in the script as follows: open server.com username ftp_commands bye Refer to blank line required after username credentials. Example from: FTP from batch file another reference to the same: http://newsgroups.derkeiler.com/Archive/Comp/comp.sys.ibm.as400.misc/2008-05/msg00227.html Also discussed here: archive.midrange.com/midrange-l/200601/msg00048.html "The behavior I'm observing is the same as if I didn't specify the password to login." with an answer referring to our same fix... archive.midrange.com/midrange-l/200601/msg00053.html and archive.midrange.com/midrange-l/200601/msg00065.html Note: It is my experience that FTP questions attract uncouth responses. Admittedly FTP is outdated, but many clients still have legacy systems, which they cannot upgrade or replace. The reason thereof should not be discussed here. The intention of this question is to invite a positive response. Please do not respond if you disagree with the above. If you have never encountered this same issue, please do not respond. I suspect this may be limited to FTP scripts executed from Windows machines, but have been told that this happens often and with many different servers. My specific interest is to understand what may cause this as I have a real world example of a production system suddenly requiring this as a workaround fix, after running for many years without issue. The server belongs to a third party who claims no change on their end. Server details unknown and cannot be determined. Any help or encouragement from someone who has come across the same, would be appreciated. ps. Sorry for the many words and references to painful responses, but I have asked similar questions on serverfault and elsewhere and unfortunately got back kneejerk responses to FTP and respondents debating the validity of the question. I would truly not ask, or re-post this question online if I had a better understanding of the issue. I know of people who have seen this issue, but don't know what causes it. I am wary that this question would again turn into another irrelevant discussion. Please, I ask very nicely: Please do not respond if you have not encountered a similar issue. FURTHER EDIT: Please do not suggest changing the product. The problem is not the blank line requirement. We know this fixes the issue. The problem is not being able to explain the reason for the blank line in the first place. Slight difference, but a critical point to note wrt the answering of this question.

  Read the article

• Migrating Split Access Database from one domain to another (not working, details in Q)

  - by Expo_Rob

  Some background: I'm a programmer, not a network administrator, who has been asked to migrate some accounting software (Integrated Office Accounting version 3.2) from an existing domain (OLD_NETWORK) to a new domain (NEW_NETWORK). No-body at the office knows how it works under the hood. It is a split Access 2000 database with the back-end shared and on a file server (which is also the DC) using mapped drives. The DC is NT Server 4 SP 6. The new server is server 2003. The two networks are running independently (ie: two computers on each desk). I have been able to get new computers set up on NEW_NETWORK and working with the IOA software just perfectly but for one problem: The company here uses other entirely separate databases which access the tables IOA maintains (specifically the 'customers' table) via links. To switch between these systems, you press F11 then File-Open the appropriate database and away you go (this is necessary to maintain the permissions that the IOA system uses to protect the customers table). The entire database is Access 2000, the links go to other Access databases, SQL-Server is not involved in any way, nor is a migration to SQL server likely. If I can't migrate anything over, everything will stay as it is, and the NEW_NETWORK computers will not be used. The problem: When I try and update these seperate databases (I shall call one "BANK_ACCOUNT", but the name does not matter), it says "this recordset cannot be updated". It also will sometimes not pull information out of the 'customers' table (ie: date_entered) when looking at a report of everyone who opened a bank account on a certain day (ie: today). I have tried: Giving 'everyone' full control via. shared directory permissions Giving 'everyone' full control on a file system level Checking the permissions within Access (everyone has full read/write on all tables) Copying the entire server contents from one file server to another (ie: xcopy everything) Copying the entire local client files from one computer to another, putting them in the exact same position in the file system, with the same permissons (or full control to 'everyone'). Running as an Administrator Taking one of the NEW_NETWORK computers, having it join OLD_NETWORK and run the software (direct copy from a working system with identical drive mappings), this did not work Weeping openly My Question: Is there anything else I can try? (sorry for this being so long)

  Read the article

• Bluetooth RFCOMM / SDP connection to a RS232 adapter in android

  - by ThePosey

  Hello All, I am trying to use the Bluetooth Chat sample API app that google provides to connect to a bluetooth RS232 adapter hooked up to another device. Here is the app for reference: http://developer.android.com/resources/samples/BluetoothChat/index.html And here is the spec sheet for the RS232 connector just for reference: http://serialio.com/download/Docs/BlueSnap-guide-4.77_Commands.pdf Well the problem is that when I go to connect to the device with: mmSocket.connect(); (BluetoothSocket::connect()) I always get an IOException error thrown by the connect() method. When I do a toString on the exception I get "Service discovery failed". My question is mostly what are the cases that would cause an IOException to get thrown in the connect method? I know those are in the source somewhere but I don't know exactly how the java layer that you write apps in and the C/C++ layer that contains the actual stacks interface. I know that it uses the bluez bluetooth stack which is written in C/C++ but not sure how that ties into the java layer which is what I would think is throwing the exception. Any help on pointing me to where I can try to dissect this issue would be incredible. Also just to note I am able to pair with the RS232 adapter just fine but I am never able to actually connect. Here is the logcat output for more reference: I/ActivityManager( 1018): Displayed activity com.example.android.BluetoothChat/.DeviceListActivity: 326 ms (total 326 ms) E/BluetoothService.cpp( 1018): stopDiscoveryNative: D-Bus error in StopDiscovery: org.bluez.Error.Failed (Invalid discovery session) D/BluetoothChat( 1729): onActivityResult -1 D/BluetoothChatService( 1729): connect to: 00:06:66:03:0C:51 D/BluetoothChatService( 1729): setState() STATE_LISTEN - STATE_CONNECTING E/BluetoothChat( 1729): + ON RESUME + I/BluetoothChat( 1729): MESSAGE_STATE_CHANGE: STATE_CONNECTING I/BluetoothChatService( 1729): BEGIN mConnectThread E/BluetoothService.cpp( 1018): stopDiscoveryNative: D-Bus error in StopDiscovery: org.bluez.Error.Failed (Invalid discovery session) E/BluetoothEventLoop.cpp( 1018): event_filter: Received signal org.bluez.Device:PropertyChanged from /org/bluez/1498/hci0/dev_00_06_66_03_0C_51 I/BluetoothChatService( 1729): CONNECTION FAIL TOSTRING: java.io.IOException: Service discovery failed D/BluetoothChatService( 1729): setState() STATE_CONNECTING - STATE_LISTEN D/BluetoothChatService( 1729): start D/BluetoothChatService( 1729): setState() STATE_LISTEN - STATE_LISTEN I/BluetoothChat( 1729): MESSAGE_STATE_CHANGE: STATE_LISTEN V/BluetoothEventRedirector( 1080): Received android.bleutooth.device.action.UUID I/NotificationService( 1018): enqueueToast pkg=com.example.android.BluetoothChat callback=android.app.ITransientNotification$Stub$Proxy@446327c8 duration=0 I/BluetoothChat( 1729): MESSAGE_STATE_CHANGE: STATE_LISTEN E/BluetoothEventLoop.cpp( 1018): event_filter: Received signal org.bluez.Device:PropertyChanged from /org/bluez/1498/hci0/dev_00_06_66_03_0C_51 V/BluetoothEventRedirector( 1080): Received android.bleutooth.device.action.UUID The device I'm trying to connect to is the 00:06:66:03:0C:51 which I can scan for and apparently pair with just fine. The below is merged from a similar question which was successfully resolved by the selected answer here: How can one connect to an rfcomm device other than another phone in Android? The Android API provides examples of using listenUsingRfcommWithServiceRecord() to set up a socket and createRfcommSocketToServiceRecord() to connect to that socket. I'm trying to connect to an embedded device with a BlueSMiRF Gold chip. My working Python code (using the PyBluez library), which I'd like to port to Android, is as follows: sock = bluetooth.BluetoothSocket(proto=bluetooth.RFCOMM) sock.connect((device_addr, 1)) return sock.makefile() ...so the service to connect to is simply defined as channel 1, without any SDP lookup. As the only documented mechanism I see in the Android API does SDP lookup of a UUID, I'm slightly at a loss. Using "sdptool browse" from my Linux host comes up empty, so I surmise that the chip in question simply lacks SDP support.

  Read the article

• ASP.NET 3.5/C# Menu Control in Master Page fails to use CSS styles

  - by Shaun

  I'm working on a web application that uses ASP.NET 3.5 and C#. Structurally, I have a master page with a menu control on it. The control serves as my navigation, and it gets its items from a SiteMapDataSource control and a corresponding Web.sitemap file. The problem is that some styles do not render properly when you specify the CssClass property. More specifically, the selected and hover styles don't respond to css styles. Consider the code below: <%@ Master Language="C#" AutoEventWireup="true" CodeFile="Site.master.cs" Inherits="Site" %> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.or/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> <head runat="server"> <title>A webpage</title> </head> <body> <form id="form1" runat="server"> <div id="page"> <asp:Menu ID="navMenu" Orientation="Horizontal" StaticMenuStyle-CssClass="staticMenu" StaticMenuItemStyle-CssClass="staticMenuItem" StaticSelectedStyle-CssClass="staticSelectedItem" StaticHoverStyle-CssClass="staticHoverItem" runat="server"> </asp:Menu> <asp:SiteMapDataSource ID="srcSiteMap" runat="server" ShowStartingNode="false" /> <br /> <asp:ContentPlaceHolder id="ContentPlaceHolder1" runat="server"> </asp:ContentPlaceHolder> </div> </form> </body> </html> Suppose I had a corresponding .css file with the following: .staticMenuItem { background-color:Red; } .staticSelectedItem { background-color:Green; } .staticHoverItem { background-color:Blue; } What will happen is that my item backgrounds will properly be red, but my selected item will not be green and the item I'm hovering my mouse over will not be blue. This seems true regardless of whether or not I include the style in the head of the master page or in an external file in default theme as specified in the web.config file. If I specify the styles in the asp.net xml like so: <asp:Menu ID="navMenu" Orientation="Horizontal" runat="server"> <StaticSelectedStyle BackColor="Green" Font-Underline="True" Font-Bold="True" /> <StaticHoverStyle BackColor="Gray" /> </asp:Menu> It appears to work properly in Firefox, but the style is never embedded in the html in Internet Explorer. Odd. Does anybody have any insight into what is causing this problem and how to neatly work around it? I'm aware I might be able to programmically determine the current page and select the corresponding menu item manually so it receives the proper style class, but before I resort to hacking C# and Javascript together to fix this functionality, I'm open to ideas. Thanks!

  Read the article

• Errors rose when a Netbean Maven Project tries to run

  - by Zakaria

  Hi everybody, I installed NetBeans 6.8 on Vista and and I'm trying to execute a simple Maven Project. When I ran the project, I got this set of errors: WARNING: You are running embedded Maven builds, some build may fail due to incompatibilities with latest Maven release. To set Maven instance to use for building, click here. Scanning for projects... [#process-resources] [resources:resources] Using default encoding to copy filtered resources. [#compile] [compiler:compile] Nothing to compile - all classes are up to date [exec:exec] [EL Info]: 2010-04-04 18:22:54.907--ServerSession(15532856)--EclipseLink, version: Eclipse Persistence Services - 2.0.0.v20091127-r5931 [EL Severe]: 2010-04-04 18:22:54.929--ServerSession(15532856)--Local Exception Stack: Exception [EclipseLink-4003] (Eclipse Persistence Services - 2.0.0.v20091127-r5931): org.eclipse.persistence.exceptions.DatabaseException Exception in thread "main" javax.persistence.PersistenceException: Exception [EclipseLink-4003] (Eclipse Persistence Services - 2.0.0.v20091127-r5931): org.eclipse.persistence.exceptions.DatabaseException Exception Description: Configuration error. Class [org.apache.derby.jdbc.ClientDriver] not found. Exception Description: Configuration error. Class [org.apache.derby.jdbc.ClientDriver] not found. at org.eclipse.persistence.exceptions.DatabaseException.configurationErrorClassNotFound(DatabaseException.java:82) at org.eclipse.persistence.sessions.DefaultConnector.loadDriverClass(DefaultConnector.java:267) at org.eclipse.persistence.sessions.DefaultConnector.connect(DefaultConnector.java:85) at org.eclipse.persistence.internal.jpa.EntityManagerSetupImpl.deploy(EntityManagerSetupImpl.java:392) at org.eclipse.persistence.sessions.DatasourceLogin.connectToDatasource(DatasourceLogin.java:162) at org.eclipse.persistence.internal.jpa.EntityManagerFactoryImpl.getServerSession(EntityManagerFactoryImpl.java:151) at org.eclipse.persistence.internal.sessions.DatabaseSessionImpl.loginAndDetectDatasource(DatabaseSessionImpl.java:584) at org.eclipse.persistence.internal.jpa.EntityManagerFactoryImpl.createEntityManagerImpl(EntityManagerFactoryImpl.java:207) at org.eclipse.persistence.internal.jpa.EntityManagerFactoryProvider.login(EntityManagerFactoryProvider.java:228) at org.eclipse.persistence.internal.jpa.EntityManagerFactoryImpl.createEntityManager(EntityManagerFactoryImpl.java:195) at com.mycompany.chapter2_ex1.Main.main(Main.java:31) Caused by: Exception [EclipseLink-4003] (Eclipse Persistence Services - 2.0.0.v20091127-r5931): org.eclipse.persistence.exceptions.DatabaseException Exception Description: Configuration error. Class [org.apache.derby.jdbc.ClientDriver] not found. at org.eclipse.persistence.internal.jpa.EntityManagerSetupImpl.deploy(EntityManagerSetupImpl.java:368) at org.eclipse.persistence.exceptions.DatabaseException.configurationErrorClassNotFound(DatabaseException.java:82) at org.eclipse.persistence.internal.jpa.EntityManagerFactoryImpl.getServerSession(EntityManagerFactoryImpl.java:151) at org.eclipse.persistence.sessions.DefaultConnector.loadDriverClass(DefaultConnector.java:267) at org.eclipse.persistence.internal.jpa.EntityManagerFactoryImpl.createEntityManagerImpl(EntityManagerFactoryImpl.java:207) at org.eclipse.persistence.sessions.DefaultConnector.connect(DefaultConnector.java:85) at org.eclipse.persistence.internal.jpa.EntityManagerFactoryImpl.createEntityManager(EntityManagerFactoryImpl.java:195) at org.eclipse.persistence.sessions.DatasourceLogin.connectToDatasource(DatasourceLogin.java:162) at com.mycompany.chapter2_ex1.Main.main(Main.java:31) at org.eclipse.persistence.internal.sessions.DatabaseSessionImpl.loginAndDetectDatasource(DatabaseSessionImpl.java:584) at org.eclipse.persistence.internal.jpa.EntityManagerFactoryProvider.login(EntityManagerFactoryProvider.java:228) at org.eclipse.persistence.internal.jpa.EntityManagerSetupImpl.deploy(EntityManagerSetupImpl.java:368) ... 4 more [ERROR]The following mojo encountered an error while executing: [ERROR]Group-Id: org.codehaus.mojo [ERROR]Artifact-Id: exec-maven-plugin [ERROR]Version: 1.1.1 [ERROR]Mojo: exec [ERROR]brought in via: Direct invocation [ERROR]While building project: [ERROR]Group-Id: com.mycompany [ERROR]Artifact-Id: chapter2_ex1 [ERROR]Version: 1.0-SNAPSHOT [ERROR]From file: C:\Users\Charlotte\Documents\NetBeansProjects\chapter2_ex1\pom.xml [ERROR]Reason: Result of cmd.exe /X /C ""C:\Program Files\Java\jdk1.6.0_11\bin\java.exe" -classpath C:\Users\Charlotte\Documents\NetBeansProjects\chapter2_ex1\target\classes;C:\Users\Charlotte\.m2\repository\javax\persistence\persistence-api\1.0\persistence-api-1.0.jar;C:\Users\Charlotte\.m2\repository\org\eclipse\persistence\javax.persistence\2.0.0\javax.persistence-2.0.0.jar;C:\Users\Charlotte\.m2\repository\org\eclipse\persistence\eclipselink\2.0.0-RC1\eclipselink-2.0.0-RC1.jar com.mycompany.chapter2_ex1.Main" execution is: '1'. ------------------------------------------------------------------------ For more information, run with the -e flag ------------------------------------------------------------------------ BUILD FAILED ------------------------------------------------------------------------ Total time: 3 seconds Finished at: Sun Apr 04 18:22:55 CEST 2010 Final Memory: 47M/94M ------------------------------------------------------------------------ Theses exceptions rose even if I can run the database by using the console (ij) and when I connect the Database, no errors are showing. Can you help me please? Thank you very much. Regards.

  Read the article

• "Access is denied" JavaScript error when trying to access the document object of a programmatically-

  - by Bungle

  I have project in which I need to create an <iframe> element using JavaScript and append it to the DOM. After that, I need to insert some content into the <iframe>. It's a widget that will be embedded in third-party websites. I don't set the "src" attribute of the <iframe> since I don't want to load a page; rather, it is used to isolate/sandbox the content that I insert into it so that I don't run into CSS or JavaScript conflicts with the parent page. I'm using JSONP to load some HTML content from a server and insert it in this <iframe>. I have this working fine, with one serious exception - if the document.domain property is set in the parent page (which it may be in certain environments in which this widget is deployed), Internet Explorer (probably all versions, but I've confirmed in 6, 7, and 8) gives me an "Access is denied" error when I try to access the document object of this <iframe> I've created. It doesn't happen in any other browsers I've tested in (all major modern ones). This makes some sense, since I'm aware that Internet Explorer requires you to set the document.domain of all windows/frames that will communicate with each other to the same value. However, I'm not aware of any way to set this value on a document that I can't access. Is anyone aware of a way to do this - somehow set the document.domain property of this dynamically created <iframe>? Or am I not looking at it from the right angle - is there another way to achieve what I'm going for without running into this problem? I do need to use an <iframe> in any case, as the isolated/sandboxed window is crucial to the functionality of this widget. Here's my test code: <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> <title>Document.domain Test</title> <script type="text/javascript"> document.domain = 'onespot.com'; // set the page's document.domain </script> </head> <body> <p>This is a paragraph above the &lt;iframe&gt;.</p> <div id="placeholder"></div> <p>This is a paragraph below the &lt;iframe&gt;.</p> <script type="text/javascript"> var iframe = document.createElement('iframe'), doc; // create <iframe> element document.getElementById('placeholder').appendChild(iframe); // append <iframe> element to the placeholder element setTimeout(function() { // set a timeout to give browsers a chance to recognize the <iframe> doc = iframe.contentWindow || iframe.contentDocument; // get a handle on the <iframe> document alert(doc); if (doc.document) { // HEREIN LIES THE PROBLEM doc = doc.document; } doc.body.innerHTML = '<h1>Hello!</h1>'; // add an element }, 10); </script> </body> </html> I've hosted it at: http://troy.onespot.com/static/access_denied.html As you'll see if you load this page in IE, at the point that I call alert(), I do have a handle on the window object of the <iframe>; I just can't get any deeper, into its document object. Thanks very much for any help or suggestions! I'll be indebted to whomever can help me find a solution to this.

  Read the article

• How to create an array of User Objects in Powerbuilder?

  - by TomatoSandwich

  The application has many different windows. One is a single 'row' window, which relates to a single row of data in a table, say 'Order'. Another is a 'multiple row' datawindow, where each row in the datawindow relates to a row in 'Order', used for spreadsheet-like data entry Functionality extentions have create a detail table, say 'Suppliers', where an order may require multiple suppliers to fill the order. Normally, suppliers are not required, because they are already in the warehouse (0), or there may need to be an order to a supplier to complete an order (1), or multiple suppliers may need to be contacted (more than one). As a single order is entered, once the items are entered, a User Object is populated depending on the status of the items in the warehouse. If required, this creates a 1-to-many relationship between the order and the "backorder". In the PB side, there is a single object uo_backorder which is created on the window, and is referenced by the window depending on the command (button popup, save, etc) I have been tasked to create the 'backorder' functionality on the spreadsheet-line window. Previously the default options for backorders were used when orders were created from the multiple-row window. A workaround already exists where unconfirmed orders could be opened in the single-row window, and the backorder information manipulated there. However, the userbase wants this functionality on the one window. Since the functionality of uo_backorder already exists, I assumed I could just copy the code from the single-order window, but create an array of uo_backorder objects to cope with multiple rows. I tried the following: forward .. type uo_backorder from popupdwpb within w_order_conv end type end forward global type w_order_conv from singleform .. uo_backorder uo_backorder end type type variables .. uo_backorder iuo_backorders[] end variables .. public function boolean iuo_backorders(); .. long ll_count ll_count = UpperBound(iuo_backorders[]) iuo_backorders[ll_count+1] = uo_backorder //THIS ISN'T RIGHT lb_ok = iuo_backorders[ll_count+1].init('w_backorder_popup', '', '', '', 'd_backorder_popup', sqlca, useTransObj()) return lb_ok end function .. <utility functions> .. type uo_backorder from popupdwpb within w_order_conv integer x = 28 integer y = 28 integer width ... end type on uo_backorder.destroy call popupdwpb::destroy end on The issue I face now is that the code commented "THIS ISN'T RIGHT" isn't correct. It is associating the visual object placed on the face of the main window to each array cell, so anytime I reference the array cell object it's actually referencing the one original object, not the new instances that I (thought) I was creating. If I change the code iuo_backorders[ll_count+1] = create uo_backorder the code doesn't run, saying that it failed to initalize the popup window. I think this is related to the class being called the same thing as the instance. What I want to end up with is an array of uo_backorder objects that I can associate to each row of my datawindow (first row = first cell, etc). I think the issue lays in the fact it's a visual object, and I can't seem to get the window to run without adding a dummy object on the face of the window (functionality from the original single-row window). Since it's a VISUAL object, does the object indeed need to be embedded on the windowface for the window to know what object I'm talking about? If so, how does one create multiple windowface objects (one to many, depending on when a row is added)? Don't hesitate to inquire regarding any more information this issue may require from myself. I have no idea what is 'standard' or 'default' in PB, or what is custom and needs more explaining.

  Read the article

• How to pre-load all deployed assemblies for an AppDomain

  - by Andras Zoltan

  Given an App Domain, there are many different locations that Fusion (the .Net assembly loader) will probe for a given assembly. Obviously, we take this functionality for granted and, since the probing appears to be embedded within the .Net runtime (Assembly._nLoad internal method seems to be the entry-point when Reflect-Loading - and I assume that implicit loading is probably covered by the same underlying algorithm), as developers we don't seem to be able to gain access to those search paths. My problem is that I have a component that does a lot of dynamic type resolution, and which needs to be able to ensure that all user-deployed assemblies for a given AppDomain are pre-loaded before it starts its work. Yes, it slows down startup - but the benefits we get from this component totally outweight this. The basic loading algorithm I've already written is as follows. It deep-scans a set of folders for any .dll (.exes are being excluded at the moment), and uses Assembly.LoadFrom to load the dll if it's AssemblyName cannot be found in the set of assemblies already loaded into the AppDomain (this is implemented inefficiently, but it can be optimized later): void PreLoad(IEnumerable<string> paths) { foreach(path p in paths) { PreLoad(p); } } void PreLoad(string p) { //all try/catch blocks are elided for brevity string[] files = null; files = Directory.GetFiles(p, "*.dll", SearchOption.AllDirectories); AssemblyName a = null; foreach (var s in files) { a = AssemblyName.GetAssemblyName(s); if (!AppDomain.CurrentDomain.GetAssemblies().Any( assembly => AssemblyName.ReferenceMatchesDefinition( assembly.GetName(), a))) Assembly.LoadFrom(s); } } LoadFrom is used because I've found that using Load() can lead to duplicate assemblies being loaded by Fusion if, when it probes for it, it doesn't find one loaded from where it expects to find it. So, with this in place, all I now have to do is to get a list in precedence order (highest to lowest) of the search paths that Fusion is going to be using when it searches for an assembly. Then I can simply iterate through them. The GAC is irrelevant for this, and I'm not interested in any environment-driven fixed paths that Fusion might use - only those paths that can be gleaned from the AppDomain which contain assemblies expressly deployed for the app. My first iteration of this simply used AppDomain.BaseDirectory. This works for services, form apps and console apps. It doesn't work for an Asp.Net website, however, since there are at least two main locations - the AppDomain.DynamicDirectory (where Asp.Net places it's dynamically generated page classes and any assemblies that the Aspx page code references), and then the site's Bin folder - which can be discovered from the AppDomain.SetupInformation.PrivateBinPath property. So I now have working code for the most basic types of apps now (Sql Server-hosted AppDomains are another story since the filesystem is virtualised) - but I came across an interesting issue a couple of days ago where this code simply doesn't work: the nUnit test runner. This uses both Shadow Copying (so my algorithm would need to be discovering and loading them from the shadow-copy drop folder, not from the bin folder) and it sets up the PrivateBinPath as being relative to the base directory. And of course there are loads of other hosting scenarios that I probably haven't considered; but which must be valid because otherwise Fusion would choke on loading the assemblies. I want to stop feeling around and introducing hack upon hack to accommodate these new scenarios as they crop up - what I want is, given an AppDomain and its setup information, the ability to produce this list of Folders that I should scan in order to pick up all the DLLs that are going to be loaded; regardless of how the AppDomain is setup. If Fusion can see them as all the same, then so should my code. Of course, I might have to alter the algorithm if .Net changes its internals - that's just a cross I'll have to bear. Equally, I'm happy to consider SQL Server and any other similar environments as edge-cases that remain unsupported for now. Any ideas!?

  Read the article

• IE attachEvent on object tag causes memory corruption

  - by larswa

  I've an ActiveX Control within an embedded IE8 HTML page that has the following event MessageReceived([in] BSTR srcWindowId, [in] BSTR json). On Windows the event is registered with OCX.attachEvent("MessageReceived", onMessageReceivedFunc). Following code fires the event in the HTML page. HRESULT Fire_MessageReceived(BSTR id, BSTR json) { CComVariant varResult; T* pT = static_cast<T*>(this); int nConnectionIndex; CComVariant* pvars = new CComVariant[2]; int nConnections = m_vec.GetSize(); for (nConnectionIndex = 0; nConnectionIndex < nConnections; nConnectionIndex++) { pT->Lock(); CComPtr<IUnknown> sp = m_vec.GetAt(nConnectionIndex); pT->Unlock(); IDispatch* pDispatch = reinterpret_cast<IDispatch*>(sp.p); if (pDispatch != NULL) { VariantClear(&varResult); pvars[1] = id; pvars[0] = json; DISPPARAMS disp = { pvars, NULL, 2, 0 }; pDispatch->Invoke(0x1, IID_NULL, LOCALE_USER_DEFAULT, DISPATCH_METHOD, &disp, &varResult, NULL, NULL); } } delete[] pvars; // -> Memory Corruption here! return varResult.scode; } After I enabled gflags.exe with application verifier, the following strange behaviour occur: After Invoke() that is executing the JavaScript callback, the BSTR from pvars[1] is copied to pvars[0] for some unknown reason!? The delete[] of pvars causes a double free of the same string then which ends in a heap corruption. Does anybody has an idea whats going on here? Is this a IE bug or is there a trick within the OCX Implementation that I'm missing? If I use the tag like: <script for="OCX" event="MessageReceived(id, json)" language="JavaScript" type="text/javascript"> window.onMessageReceivedFunc(windowId, json); </script> ... the strange copy operation does not occur. The following code also seem to be ok due to the fact that the caller of Fire_MessageReceived() is responsible for freeing the BSTRs. HRESULT Fire_MessageReceived(BSTR srcWindowId, BSTR json) { CComVariant varResult; T* pT = static_cast<T*>(this); int nConnectionIndex; VARIANT pvars[2]; int nConnections = m_vec.GetSize(); for (nConnectionIndex = 0; nConnectionIndex < nConnections; nConnectionIndex++) { pT->Lock(); CComPtr<IUnknown> sp = m_vec.GetAt(nConnectionIndex); pT->Unlock(); IDispatch* pDispatch = reinterpret_cast<IDispatch*>(sp.p); if (pDispatch != NULL) { VariantClear(&varResult); pvars[1].vt = VT_BSTR; pvars[1].bstrVal = srcWindowId; pvars[0].vt = VT_BSTR; pvars[0].bstrVal = json; DISPPARAMS disp = { pvars, NULL, 2, 0 }; pDispatch->Invoke(0x1, IID_NULL, LOCALE_USER_DEFAULT, DISPATCH_METHOD, &disp, &varResult, NULL, NULL); } } delete[] pvars; return varResult.scode; } Thanks!

  Read the article

• Is possible to make mt.exe embed manifest files correctly in Visual Studio 2008?

  - by Sorin Sbarnea

  I found that mt.exe fails to correctly create and embed manifest files into executables when run inside a VCPROJ. For example the same executable load well on Windows 7 but failed to load on Windows XP. The manifest was embedded and correct. After I spend lots of hours searching for possible reasons and solution I modified the project settings to generate the manifest outside the exe file. Now it works on both systems. Here are the examples for debug builds. With embed disabled: <?xml version="1.0" encoding="UTF-8" standalone="yes"?> <assembly xmlns="urn:schemas-microsoft-com:asm.v1" manifestVersion="1.0"> <trustInfo xmlns="urn:schemas-microsoft-com:asm.v3"> <security> <requestedPrivileges> <requestedExecutionLevel level="asInvoker" uiAccess="false"></requestedExecutionLevel> </requestedPrivileges> </security> </trustInfo> <dependency> <dependentAssembly> <assemblyIdentity type="win32" name="Microsoft.VC90.DebugCRT" version="9.0.21022.8" processorArchitecture="x86" publicKeyToken="1fc8b3b9a1e18e3b"></assemblyIdentity> </dependentAssembly> </dependency> <dependency> <dependentAssembly> <assemblyIdentity type="win32" name="Microsoft.VC90.DebugMFC" version="9.0.21022.8" processorArchitecture="x86" publicKeyToken="1fc8b3b9a1e18e3b"></assemblyIdentity> </dependentAssembly> </dependency> </assembly> This is with embed enabled: <?xml version="1.0" encoding="UTF-8" standalone="yes" ?> <assembly xmlns="urn:schemas-microsoft-com:asm.v1" manifestVersion="1.0"> <trustInfo xmlns="urn:schemas-microsoft-com:asm.v3"> <security> <requestedPrivileges> <requestedExecutionLevel level="asInvoker" uiAccess="false" /> </requestedPrivileges> </security> </trustInfo> <dependency> <dependentAssembly> <assemblyIdentity type="win32" name="Microsoft.VC90.DebugCRT" version="9.0.21022.8" processorArchitecture="x86" publicKeyToken="1fc8b3b9a1e18e3b" /> </dependentAssembly> </dependency> <dependency> <dependentAssembly> <assemblyIdentity type="win32" name="Microsoft.VC90.DebugMFC" version="9.0.21022.8" processorArchitecture="x86" publicKeyToken="1fc8b3b9a1e18e3b" /> </dependentAssembly> </dependency> <dependency> <dependentAssembly> <assemblyIdentity type="win32" name="Microsoft.Windows.Common-Controls" version="6.0.0.0" processorArchitecture="x86" publicKeyToken="6595b64144ccf1df" language="*" /> </dependentAssembly> </dependency> </assembly> If you compare them the second one adds common controls (I don't know from where) and also it is a small difference with the syntax of requestedExecutionLevel tag.

  Read the article

• NHibernate unmapped class exception

  - by John Prideaux

  I am trying to implement a one-to-many relationship using NHibernate 2.1.2 but keep getting "Association references unmapped class" exceptions. I have verified that my hbm.xml files are embedded resource. Here are my classes and mappings. Any ideas? public class OrderStatus { public virtual decimal MainCommit { get; set; } public virtual decimal CommitNumber { get; set; } public virtual string InvoiceNumber { get; set; } public virtual string ShipTo { get; set; } public virtual string CustomerOrderNumber { get; set; } public virtual string Station { get; set; } public virtual DateTime RequestedShipDate { get; set; } public virtual decimal EstimatedValue { get; set; } public virtual decimal EstimatedWeight { get; set; } public virtual string Customer { get; set; } public virtual DateTime InvoiceDate { get; set; } public virtual ICollection<Promise> Promises { get; set; } } <class name="AladdinDb.Models.OrderStatus, AladdinDb" table="vorder_status"> <id name="CommitNumber" type="decimal" column="commit_no"> <generator class="assigned"> <param name="property"> Plan </param> </generator> </id> <property name="MainCommit" column="main_commit" type="decimal" /> <property name="InvoiceNumber" column="invoice_no" type="string" /> <property name="ShipTo" column="ship_to" type ="string"/> <property name="CustomerOrderNumber" column="cust_order_no" type="string" /> <property name="Station" column="station" type="string" /> <property name="RequestedShipDate" column="req_ship_date" type="DateTime" /> <property name="EstimatedValue" column="estimated_value" type="decimal"/> <property name="EstimatedWeight" column="estimated_weight" type="decimal" /> <property name="Customer" column="customer" type="string" /> <property name="InvoiceDate" column="invoice_date" /> <set name="Promises"> <key column="commit_no"></key> <one-to-many class="Promise" /> </set> </class> public class Promise { public virtual decimal CommitNumber { get; set; } public virtual DateTime PromiseDate { get; set; } public virtual string WhoAsked { get; set; } public virtual string WhoGave { get; set; } public virtual string Iffy { get; set; } } <class name="AladdinDb.Models.Promise, AladdinDb" table="promise"> <id name="CommitNumber" type="decimal" column="commit_no"> <generator class="assigned" /> </id> <property name="PromiseDate" column="promise_date" /> <property name="WhoAsked" column="who_asked" /> <property name="WhoGave" column="who_gave" /> <property name="Iffy" column="iffy" /> </class>

  Read the article

< Previous Page | 357 358 359 360 361 362 363 364 365 366 367 368  | Next Page >