Yaron Gilor
Yaron Gilor
A similar error was reproduced 2 hours later: ``` disrupt_load_and_stream | longevity-tls-1tb-7d-2022-2-db-node-4b61603c-3 | Failed | 2022-09-16 04:22:00 | 2022-09-16 04:26:34 -- | -- | -- | -- | -- Nemesis...
Tested in: https://jenkins.scylladb.com/job/scylla-staging/job/Longevity_yaron/job/rolling-upgrade-debian10-test-large-partitions/9/ ==> looks like it runs ok without s-b stress. next step is try pre-creating the s-b table in advance and retest.
Reproduced in 5.0 as well, so we started using a workaround by setting Gemini parameter `--max-partition-keys 4 ` (instead of default 6). https://jenkins.scylladb.com/job/scylla-master/job/longevity/job/byo-longevity-test/182/ run failed. SCT reported error: ``` AssertionError:...
> @yarongilor note that if you want, you can also work around this by configuring Scylla (instead of Gemini), giving it a higher `--max-clustering-key-restrictions-per-query`. Right, if it sounds reasonable, SCT...
Reproduced in a terminate-and-replace nemesis as well (in An Alternator-TTL test: ``` (DatabaseLogEvent Severity.ERROR) period_type=one-time event_id=4d4f44a9-e39e-4f8e-9b58-e93df2718728: type=DATABASE_ERROR regex=(^ERROR|!\s*?ERR).*\[shard.*\] line_number=101059 node=alternator-ttl-4-loaders-no-lwt-sis-db-node-7da36ba4-5 2022-11-04T06:28:08+00:00 alternator-ttl-4-loaders-no-lwt-sis-db-node-7da36ba4-5 !ERR | scylla[58893]: [shard 6] storage_proxy - Failed...
@kostja , @elcallio , can you please advice - This error is reproduced with Scylla version (or git commit hash): `2023.1.0~rc3-20230321.80de75947b7a` with build-id `6e1d6cb6cac9242e7ed7bfd8b07c1fc5998281dc` [Argus](https://argus.scylladb.com/test/a37fcc95-6965-453f-b70a-9536989fbee8/runs?additionalRuns%5B%5D=cf2a1a67-b3db-4831-afb9-935398d9527a) It happens after a node...
> @yarongilor @elcallio write above that the error itself can be expected, it's merely a warning. Do you reproduce the crash? The crash should be gone by now. @kostja ,...
> Is that the right repo to report this? Yes. At least it looks so, according to previously existing similar issues.
ok, thanks, i see. @fgelcer , @fruch , should we switch back to the previously suggested option of using `tlp-stress` instead?
> I would recommend testing the upstream cassandra-stress from 4.1, if it's kind of working there, backporting might be easy. > > Getting familiar with tlp-stress might also be a...