13 Aug
2020
13 Aug
'20
3:56 p.m.
This still isn't enough. Please let me know if there are test logs of
yours that can be removed.
It might be helpful to just delete everything from 2019 or potentially
keeping only 6 months of logs total.
On 8/12/20 9:57 AM, David Galloway wrote:
Hi all,
Due to increases in amount of testing and length of logs, the (Long
Running) Ceph cluster in the Sepia lab has been reaching 95-98% capacity
over the past few days. Since almost everything else got deleted on the
cluster a few months ago, I need to reduce the amount of test logs we
keep on hand.
Currently we:
- Keep 14 days of passed job logs
- Compress failed job logs older than 30 days
- Delete failed job logs older than 365 days
We will now be deleting failed job logs older than 300 days. We may be
able to increase the cluster's capacity with the purchase of additional
hardware which I will discuss with the appropriate stakeholders.