Sébastien Beyou

The computation of `f_bavail` displayed by `df` is computed by the kernel in [fs/btrfs/super.c, function `btrfs_statfs`](https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/fs/btrfs/super.c?h=v5.12.10#n2241), so strictly speaking it relates to the kernel only. If there is an issue,...

I think I understand the problem, and this is roughly what Andrei Borzenkov described. In the following of my comment, "largest/smallest devices" should be understand as "devices where it remains...

Well, the situation is a bit more vicious: my proposition would work better but is not optimal in the case where it remains equally as much space on all devices....

I don’t have RAID1 on the data (only on the metadata), I know it could be a partial workaround for this case.

Many thanks for your answer @Zygo ! I read it a long time ago, I should have thank you before. BTW I had no issue since the file was deleted,...

I fixed most of the remaining errors in Travis, it remains, for 1.35, 2 failures and 2 warnings for 21782 assertions (there were yesterday 27 errors, 2 failures, 5 warnings...

A work hypothesis is that the failing step uses a connection from the ConnectionPool initialized at the beginning of the script and this connection was dropped the many hours later.

Thanks @jaideraf, I was able to setup ElasticStore without issue on a small wiki with this configuration, but the error still occur on the bigger wiki with your configuration (I...

I am retrying using a different, [more reliable](https://github.com/elastic/elasticsearch-php/blob/master/docs/connection-pool.asciidoc), ConnectionPool: StaticConnectionPool instead of StaticNoPingConnectionPool. I did so by directly modifying SMW (I didn’t see a parameter for that) in the file...

Update: it fails even with StaticConnectionPool, and I retried on a standalone server (instead of the managed infrastructure) with the same versions and with the Elasticsearch on localhost and it...