nfsd.service: Order after mxmount.service #217
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Currently there is no order between nfsd.service and mxmount.service. If mxmount is slow (e.g. when file systems have to process their journals after a server crash), nfsd startup might execute `exportfs -ra` while the filesystems are not yet mounted, thereby exporting the unmounted mountpoints. This can result in mount failures on nfs clients or in "stale NFS handle" errors on clients, which had the filesystems mounted before the crash. mxmount.service uses `mxmount --noexport` so there is no reexport triggered, after the filesystems are mounted. nfsd.services executes additional `exportfs -ra` commands 10, 20 and 30 seconds after the nfsd startup, but 30 seconds might not be enough time to mount all filesystems after the crash of a fileserver. These errors can persist. They are partly resolved by a manual `exportfs -ra` after a longer time, making the file systems available for mounting. However, the "stale NFS handle" problem might still be visible on clients which picked up the now covered inodes of the mountpoints. Order nfsd.service after mxmount.service so that we don't export the mountpoints.
donald
force-pushed
the
nfsd-after-mxmount
branch
from
00d49f7
to
2c7ad52
Compare
|
Tested on dose (with artificial |
Sign in
to join this conversation on GitHub.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Currently there is no order between nfsd.service and mxmount.service. If
mxmount is slow (e.g. when file systems have to process their journals
after a server crash), nfsd startup might execute
exportfs -rawhilethe filesystems are not yet mounted, thereby exporting the unmounted
mountpoints.
This can result in mount failures on nfs clients or in "stale NFS
handle" errors on clients, which had the filesystems mounted before the
crash.
mxmount.service uses
mxmount --noexportso there is no reexporttriggered, after the filesystems are mounted. nfsd.services executes
additional
exportfs -racommands 10, 20 and 30 seconds after the nfsdstartup, but 30 seconds might not be enough time to mount all
filesystems after the crash of a fileserver.
These errors can persist. They are partly resolved by a manual
exportfs -raafter a longer time, making the file systems available formounting. However, the "stale NFS handle" problem might still be visible
on clients which picked up the now covered inodes of the mountpoints.
Order nfsd.service after mxmount.service so that we don't export the
mountpoints.