Harbor replication does not give any reason of failed event

[rajatrj16]

There is no proper error message in harbor Why replication fails randomly?

Upon going to logs in failed event there are details which is not clear why the error is {"errors":[{"code":"NOT_FOUND","message":"{\"code\":10010,\"message\":\"object is not found\",\"details\":\"97833e43ff5b2a1e3fa3d790\"}"}]}

[AMC-7]

What does it say in the replication logs?

[rajatrj16]

@AMC-7 This is the error comes in the replication logs {"errors":[{"code":"NOT_FOUND","message":"{\"code\":10010,\"message\":\"object is not found\",\"details\":\"97833e43ff5b2a1e3fa3d790\"}"}]}

[cataglyphis]

same issue in 2.9.4

[nikokirilenko]

encountered this on v2.10.1 as well

[Doriangaensslen]

Im not sure, if you have the same problem or not. But the error is exactly what we received. We had all inodes used in one pod. harbor [ /harbor ]$ df -i /dev/sdc 65536 65536 0 100% /var/log/jobs

This means, there are so many small files, that all inodes are used and no new one can be written. It has the same effect, as when all space is used.