Linux Repository /tmp not enough space

[Baptiste]

Hello guys,

I've a linux repository in AWS and 43 agents backing up to it.
Last night, many job trigger :

Code: Select all

16/03/2018 05:10:43 :: Error: scp: /tmp/VeeamAgent45e208e2-df35-4165-8137-abb2c61d1b42: No space left on device

My / is full on my repository (30G) (my actual repository is fine). /tmp/ is full of files.

My question is : Can i clean it ? How does it clean itself ? what's are the requirements for space ?

Cannot find system partition size requirement in https://helpcenter.veeam.com/docs/backu ... er=95#repo
thanks,

[Baptiste]

I've dig into it, it's not /tmp/ which take space but the home directory of user :
It's full of core.21312 from 300Mo to 700Mo :
-rw------- 1 root root 782M Mar 15 10:14 core.6286
-rw------- 1 root root 264M Mar 15 10:18 core.6381
-rw------- 1 root root 737M Mar 15 11:19 core.6542

how does it works ?

[Gostev]

These are Linux dump files which are automatically generated when a process is terminated by the OS due to a fault in the program. You can just delete these files, unless you want to investigate those crashes with the vendors behind crashing apps. Thanks!

[Baptiste]

Hello Gostev,

Thanks for the quick reply. This linux server is only a Veeam repository, thus, is it Veeam who generate "fault in the program" ? Is there a way to disable those file generation ?
Plus, there is no system size requirement in the helpcenter for Veeam linux repo, any tips ?

thx

[Gostev]

It can be any system process or daemon failing, too. If you suspect Veeam in generating this faults, you should contact our support for investigation.

As for disabling dump files generation, you should refer to the documentation of your chosen Linux distribution.

Thanks!

[Baptiste]

i've deleted all the core file and now my copy jobs fail with :
16/03/2018 12:38:44 :: Error: Object '989f7ed1-18bc-4fd2-8c2b-3ad5c4a347fa' was not found

Case is open i'll ask them to investigate.

Regards,