Frank Filz comments

Results 698 comments of


                                            Frank Filz

5.6: [fd_lru] fsal_start_fd_work :RW LOCK :CRIT :Error 22, acquiring mutex

Could you try V5.9? I don't see anything in that log that helps.

Could not create export for (/cephfs) to (/)

We need more information to be able to analyze what is going on. Can you share your configs and scan the ceph logs for errors?

Could not create export for (/cephfs) to (/)

OK, closing.

IO Zero during parallel metadata operations on huge directories

I fear that dropping the lock while populating a chunk will open to more issues with contention on that chunk. I need to look at this closer. What is causing...

IO Zero during parallel metadata operations on huge directories

Hmm, is there any way to resolve that slowness? I'm not sure what is reasonable for Ganesha to do in that circumstance. readdir plus might help, but I think the...

Failed to get attrs for referral

So what actual errors are you seeing? I don't see any errors in the log (beyond a handful of errors during startup that shouldn't cause problems).

something wrong with the PNFS feature of nfs-ganesha

The pNFS code in Ganesha is very stale. There has not been active maintenance for quite some time. Additionally, both LIZARDFS and KVSFS are community contributions that have been dropped...

Assertion caused by the concurrency of lru_run and destroy_fsals.

Oops, the LRU reclaim thread should be stopped before shutting down the FSALs...

Assertion caused by the concurrency of lru_run and destroy_fsals.

Which version of Ganesha? Do you need a clean shutdown of Ganesha? This bug is in our backlog but we have a small team.

Assertion caused by the concurrency of lru_run and destroy_fsals.

If you don't need a clean shutdown, you could just SIGKILL instead of SIGTERM. The issue is in our queue, it's just a matter of finding time with a small...