-
Notifications
You must be signed in to change notification settings - Fork 2.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
compactor: fails "invalid checksum" #2128
Comments
Hi and thanks for report! We seen those errors before on some object storages and it looks like some corrupted data in the obj store itself. What this says is that one of the chunk files are broken for any of the: Is minio distributed or single node? Maybe it's eventually consistent and we need https://thanos.io/proposals/201901-read-write-operations-bucket.md/ (which in progress?) What is interesting is that all of those groups failed... Something that would be extremely helpful, would be to download any of those blocks and make it available for us for investigation. Would that be possible @Zergvl ? Also cc @krasi-georgiev as it looks our better errors for chunk reading works well (: But it would nice to understand this problem @krasi-georgiev |
yeah it looks like some partial upload or download, but is is surprising that is for more than a single block. |
Mailed you! |
Tried to clear storage, and exclude balancer to minio. Problem still happened =( . |
This issue/PR has been automatically marked as stale because it has not had recent activity. Please comment on status otherwise the issue will be closed in a week. Thank you for your contributions. |
Should be fixed with #3795. |
Thanos, Prometheus and Golang version used:
thanos, version 0.10.1
go version: go1.13.1
prometheus, version 2.15.2
Object Storage Provider:
S3 minio
What happened:
thanos compact --wait --log.level=debug --data-dir=/thanos/compact --http-address=x.x.x.x:19191 --objstore.config-file=/opt/thanos/storage.yaml --retention.resolution-raw=21d --retention.resolution-5m=90d --retention.resolution-1h=365d --consistency-delay=30m --block-sync-concurrency=20 --compact.concurrency=5
level=info ts=2020-02-13T05:40:10.336479317Z caller=prober.go:137 msg="changing probe status" status=not-healthy reason="error executing compaction: compaction failed: 3 errors: compaction failed for group 0@3615588222792942986: compact blocks [/thanos/compact/compact/0@3615588222792942986/01E09MMHZ2T5J3ZMTN4R0DJG2J /thanos/compact/compact/0@3615588222792942986/01E09VG972MW8JM3A74SM57BA9 /thanos/compact/compact/0@3615588222792942986/01E0A2C0F21KY74XA6VB43FEXK /thanos/compact/compact/0@3615588222792942986/01E0A97QQ26W046HY0A14DVM9Z]: write compaction: iterate compaction set: chunk 134217715 not found: segment doesn't include enough bytes to read the chunk - required:134217829, available:134217728; compaction failed for group 0@10694505281190012710: compact blocks [/thanos/compact/compact/0@10694505281190012710/01E09MMHXTHJWZTBX4DSVEHWWQ /thanos/compact/compact/0@10694505281190012710/01E09VG95EYKXMHZWFX8DK6MAQ /thanos/compact/compact/0@10694505281190012710/01E0A2C0DGW6BQXW8E9K4CMHZ4 /thanos/compact/compact/0@10694505281190012710/01E0A97QNS22R5Z64FYBR1W27Y]: write compaction: iterate compaction set: chunk 134217302 not found: segment doesn't include enough bytes to read the chunk - required:134218156, available:134217728; compaction failed for group 0@15871345448339717723: gather index issues for block /thanos/compact/compact/0@15871345448339717723/01DW4CKA83QVDRSYT8NBZ5545G: open index file: read TOC: read TOC: invalid checksum"
What you expected to happen:
dont fail
How to reproduce it (as minimally and precisely as possible):
Full logs to relevant components:
Anything else we need to know:
The text was updated successfully, but these errors were encountered: