default Linux tmp directory /root/tmp/VeeamBackupTemp/

[pirx]

What is the reason that Veeam uses /root/tmp/VeeamBackupTemp/ on Linux as a default tmp directory? Why not simply use /tmp that is the default in Linux for temporaray files? I know that I can change this is Veeam Linux config but I think it's a bad choice to set this as a default. In my case this resulted multiple times in 100% full / filesystem as /root/... is usually not a dedicated fs. With /tmp the chances are higher to not fill-up / and it would follow best practices for Linux. The issue happens during rescan of a SOBR with capacity tier, suddenly there are 10-20 GB of Indexs temporary stored there.


https://refspecs.linuxfoundation.org/FH ... 03s18.html

No space left on device
POSIX: Failed to write data to the file [/root/tmp/VeeamBackupTemp/VeeamIndexResync_{2d5bd027-bd38-43d5-9871-d3c6928cb0b0}.10faf9.7f87177fe700/e95d2022-c26f-592a-17b6-22cdbac15fe2.30].
Exception from server: Reconnectable protocol device was closed.

[HannesK]

Hello,
just to clarify... in which scenario do you see that? Because normally we use /tmp and /opt/veeam

Best regards,
Hannes

[pirx]

I can't check currently, but last time I noticed it was a fresh install of v11 veeam server

[pirx]

It's agent/offload related and even documented. Still I don't understand why to use a different folder or /root/tmp/VeeamBackupTemp at all.

https://www.veeam.com/kb4283

[HannesK]

Now I understand the problem. I had a discussion about that path recently. Without going into details, the path should be a path on the volume with most free space (same as database log files temporary files). I think, always using /tmp is also not the solution for everything.