I do not believe that the temp folder should be filling up this fast under normal circumstances. What types of files are in there? Can you send a screenshot of some sample directory contents?
Is this a distributed system, and this is on the secondary data engine host node?
not a distributed system. whole thing just on one VM.
On my test machine, most of the things in that directory are instantiations of TWBs and flat-file data sources. Maybe Tableau Server uses it during extract creation as well.
- If you log into Windows on the Tableau Server machine as the service account (AKA the "Run As Account) and try to delete the files, can you? I'm wondering specifically about permissions. See the page at http://onlinehelp.tableausoftware.com/current/server/en-us/help.htm#runas_confirm.htm to confirm.
- Have you ever made modifications to the reap frequency? (Probably not.)
- Are most extract refreshes completing successfully?
- What version are you running?
The best way to move forward on this issue would be to open a support case, unless the above questions illuminate the something that helps solve it.
Answering your questions...
1. Yes, I can delete using the service account. I manually deleted a whole heap of files to test this yesterday and also to recover about 60GB of space (wanted to see if I can recover space without stopping the service)
2. Never touched the reap frequency
3. most extract refreshes are completing successfully. That is until disk space runs out and they start failing.
4. I'm running 8.2.2 (8200.14.08.19.2015) 64 Bit.
I'll raise a support case. Thanks!!
Good luck on the support case!
One more question that occurred to me after the fact: If you look at reaps in the background tasks, are they all succeeding?
1 of 1 people found this helpful
they've come back already!! it's great!!
they said "the root cause of this issue is that A new .TDE file is created every time an incremental extract is run. By default, these files will expire and be cleaned up after 24 hours. This setting will make it expire at 12 hours."
and asked me to run the following command to limit it to 6 hours
tabadmin set temp.file_expiry_s 43200
Johnny, thank you for supplying the answer
Is that still the case for TSM?
I am also having this problem, with Tableau Server 2018.2.1 (tsm).
I raised a support case, but i'm afraid they will take too long to respond.
I don't believe it has anything to do with retention, because some of the files and folders are several months old.
My question is: is it ok to delete the older files (say everything more than a week old) with the server running? Or should i "tsm stop" it first?
Having same issue the folder is growing day by day to take more space. Can we have a resolution? We are on 2018.2 version
I added a step to delete these files, to our daily maintenance script. At a point in the script, I down Tableau Server, empty that directory, and bring it up again. We have been running like this for about a month now, without visible problems.
The permanent solution is to upgrade out of 2018.2. Higher versions of 2018 don’t seem to have the problem.