In case someone in the future has the same problem with an osd being to full to boot, I was able to recover the data by copying the underlying disk to a bigger disk.
dd if=source/drive of=dest/drive bs=32M status=progress
reboot the node with the source disk detached.
Resize the lvm...
Edit: Should have probably attached the Crashdump without people having to download the log file
Journalctl only shows the crashdump at also shows up at the end of the log.
Don't want to recreate the osd just yet.
Already look through the troubleshooting doc, also searched the issue tracker...
Hey,
one of my osd's wont start anymore
already tried the bitmap allocator still the same error
Crash dump:
2022-04-07T23:35:39.127+0200 7f3f1403af00 1 bluefs _allocate unable to allocate 0x400000 on bdev 1, allocator name block, allocator type bitmap, capacity 0x3a38800000, block size...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.