Broker restart loop after hard shutdown

hmm

is there something in the data directory which contains metadata in its folder name?

Nope.

__cluster_metadata-0]$ ls -ltr
total 54072
-rw-r–r–. 1 232071 232071 43 Feb 11 07:42 partition.metadata
-r--------. 1 232071 232071 13320 Jun 10 09:38 00000000000020564285-0000001218.checkpoint
[…]
-r--------. 1 232071 232071 13320 Jun 14 14:39 00000000000021290657-0000001311.checkpoint
-r--------. 1 232071 232071 13320 Jun 14 15:39 00000000000021297986-0000001313.checkpoint
-rw-r–r–. 1 232071 232071 621 Jun 14 15:39 leader-epoch-checkpoint
-rw-r–r–. 1 232071 232071 53385962 Jun 14 15:53 00000000000020558244.log
-rw-r–r–. 1 232071 232071 156 Jul 3 08:04 quorum-state
-rw-r–r–. 1 232071 232071 10 Jul 3 08:05 00000000000021299787.snapshot
-rw-r–r–. 1 232071 232071 10485756 Jul 3 08:05 00000000000020558244.timeindex
-rw-r–r–. 1 232071 232071 10485760 Jul 3 08:05 00000000000020558244.index

not cool

maybe a reinit would be the best
and remove “latest” from you prod deployments…

Hi,

Ok, will do that then, thanks for trying:)

Whats the recommendation to get the latest respective branch updates?
Or do you recommend to just do manual image selection all the time?

Thanks

Hi,

very welcome :slight_smile:

I tend to do this manually
for confluent images it’s worth to keep a look at this gh repo

1 Like

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.