IIS Worker Process

[Anonymous]

Hi, i have some kind a problem with Vault and IIS, now im running in 2018 update 2:

I found this:

But I don't know what i need to change.

I can't use Vault with Inventor/Acad and with Vault Client too.

If anyone have some idea do share with me.

Thanks

[JasonDai]

Hello Leokornowski,

If your Vault has just gone through a lot of changes, e.g. migration from an older version, or huge number of files just checked in, then possibly Vault is busy with some background work. Otherwise very possibly it would be an unexpected hang. You can reset IIS to recover: 1) run cmd as administor, 2) "iisreset".

iis logs (default location: C:\inetpub\logs\LogFiles\W3SVC1) and Vault logs (default location: C:\ProgramData\Autodesk\VaultServer\FileStore) will help on identifying the root cause.

Thanks,

Jason

[Anonymous]

Hi @JasonDai

Yes, this Vault is a new instalation and a recovery from preview backup, like 450Gb in total, after recovery i made a defrag and a reindex.
And this Vault is working for a few days, i think a month, and now started do this.
I will check all logs, and soon as possible i will post here.

Thanks for the help.

[JasonDai]

Hi Leokornowski,

You're welcome. This shouldn't be any background work since it has been good for one month. Considering it's a new installation, possibly there are some configurations pending to tweak. For example, we know if SQL remote query timeout is not 0, sever would hang as you have monitored. But we need to see the logs to have a precise conclusion.

Thanks,

Jason

[johannes.bauer]

You could also check the worker processes queue to see what is taking so long.

Select the IIS and workprocesses and see which application pool is taking most memory / cpu

[Anonymous]

Hi @JasonDai, i already have the log, but is huge, there is a list of files into that folders, and i really don't know what Im looking for.

There ismore than 2Gb of files, i'm already try to read some files, i can found nothing suspicious.

 In this link https://autode.sk/2EGOJgd have all log files.

Thanks for the help guys.

[Anonymous]

Hi @johannes.bauer, soon as i get this info i will share with you.

[JasonDai]

Hi @Anonymous,

I see 14 SQL timeout errors in vlog-20180409.txt, which is the log of the day you posted this problem. That should be relevant to the performance issue. Please follow below steps to increase the timeout limits--

1. Follow https://docs.microsoft.com/en-us/sql/database-engine/configure-windows/configure-the-remote-query-timeout-server-configuration-option to set SQL remote query timeout to 0
2. Restart SQL server.
3. Open Windows Explorer and navigate to the C:\Program Files\Autodesk\Autodesk Data Management Console [year]\Server\Web\Services directory, where [year] is the release year for the Autodesk Data Management Console.
4. Create a backup copy of the Web.config file.
5. Open the Web.config file with Notepad and look for the following line:<timeouts connection="120" defaultCommand="360" longCommand="1800"/> And give higher timeout values – I don’t have a recommended value here as it should be different according to the database size and network quality. Try with 3 times values (360, 1080, 5600) first maybe, if there are still any timeout error, increase it until the problem is gone.
6. Save the Web.config file.
7. Open a command prompt as administrator and type IISRESET.

I'm 80% sure your problem could be resolved after these timeout errors are gone. But if it persisted, please report back to us and let's do more investigation.

Thanks,

Jason

[JasonDai]

BTW, practices that helps on SQL performance, such as  defragment database, will also help on your problem. Please reference to https://knowledge.autodesk.com/search-result/caas/CloudHelp/cloudhelp/2017/ENU/Vault-Admin/files/GUI... for more information.

Thanks,

Jason

[Anonymous]

I think is the ThinClient:

I alreday made some changes in SQL timeout and web.services.

Maybe till the end of the week i have something more to tell you guys.

[johannes.bauer]

Hi,

you can check the app pool and recycle this for now.

If you want to see which calls are stuck, look at the worker processes. There are sure some in the pipe which should disappear by recylcing the app pool

[Anonymous]

I will try to adjust the times for recycling, but i dont have a clue what value i need to fill, so, i will just put some value, and start monitoring during the day.