CONFIG_[NFS|AFS]_FSCACHE needs to be turned on
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
| linux (Ubuntu) |
Medium
|
Unassigned |
Bug Description
Binary package hint: cachefilesd
The package cachefilesd is contained in lucid. However, while the official lucid kernel has CONFIG_FSCACHE on, CONFIG_NFS_FSCACHE and CONFIG_AFS_FSCACHE are off, meaning these filesystems do not have fscache support. So, cachefilesd runs, but does nothing, which is not very obvious.
If possible, future lucid kernels should have these options on. If this is not possible, cachefilesd in lucid should at least give a warning during installation that a newer kernel (linux-
Morten Kallesøe (kal) wrote : | #1 |
Launchpad Janitor (janitor) wrote : | #2 |
Status changed to 'Confirmed' because the bug affects multiple users.
Changed in cachefilesd (Ubuntu): | |
status: | New → Confirmed |
tags: | added: precise |
tags: | added: kernel-bug quantal |
Pierre Carrier (pcarrier) wrote : | #3 |
CONFIG_CIFS_FSCACHE still missing in precise.
affects: | cachefilesd (Ubuntu) → linux (Ubuntu) |
Changed in linux (Ubuntu): | |
importance: | Undecided → Medium |
status: | Confirmed → Triaged |
Leann Ogasawara (leannogasawara) wrote : | #4 |
For the Quantal 12.10 kernel, it appears that the following are already enabled:
~/ubuntu-
config.
~/ubuntu-
config.
However, CONFIG_AFS_FSCACHE still appears to be disabled at the moment:
~/ubuntu-
config.
Examining CONFIG_AFS_FSCACHE further, it appears it is marked as EXPERIMENTAL. ie:
~/ubuntu-quantal$ grep -A7 "config AFS_FSCACHE" fs/afs/Kconfig config AFS_FSCACHE
bool "Provide AFS client caching support (EXPERIMENTAL)"
depends on EXPERIMENTAL
depends on AFS_FS=m && FSCACHE || AFS_FS=y && FSCACHE=y
help
Say Y here if you want AFS data to be cached locally on disk through
the generic filesystem cache manager
Our default configuration policy is to disable EXPERIMENTAL options until there has been a specific request to enable them and the requested option would result in minimal risk of regression/negative impact. I've subsequently applied the attached patch which enables CONFIG_
Changed in linux (Ubuntu): | |
status: | Triaged → Fix Committed |
Pierre Carrier (pcarrier) wrote : | #5 |
I haven't run any benchmarks yet, but use resources from a CIFS server across the Internet, sometimes through a VPNish solution like sshuttle.
I have little doubts that the performance improvement could be quite significant for certain loads, and the risk for a mount point without the "fsc" mount option seems very low (looking at CIFS_MOUNT_FSCACHE in Ubuntu-
So great to hear that we'll get this option in Quantal.
Any chance it could also be considered for a latter upgrade of precise?
tags: | added: patch |
Launchpad Janitor (janitor) wrote : | #6 |
This bug was fixed in the package linux - 3.5.0-10.10
---------------
linux (3.5.0-10.10) quantal-proposed; urgency=low
[ Andy Whitcroft ]
* rebase to v3.5.1
[ Leann Ogasawara ]
* [Config] Enable CONFIG_
- LP: #728977
[ Tim Gardner ]
* SAUCE: firmware: Remove emi62 files duplicated in linux-firmware
* SAUCE: firmware: Remove tehuti files duplicated in linux-firmware
[ Upstream Kernel Changes ]
* overlayfs: copy up i_uid/i_gid from the underlying inode
- LP: #944386
* hwmon: (applesmc) Shorten minimum wait time
- LP: #1034449
* hwmon: (applesmc) Decode and act on read/write status codes
- LP: #1034449
[ Upstream Kernel Changes ]
* rebase to v3.5.1
- LP: #1026953
- LP: #1025377
-- Leann Ogasawara <email address hidden> Sun, 12 Aug 2012 13:16:43 -0700
Changed in linux (Ubuntu): | |
status: | Fix Committed → Fix Released |
Adam Conrad (adconrad) wrote : Update Released | #7 |
The verification of this Stable Release Update has completed successfully and the package has now been released to -updates. Subsequently, the Ubuntu Stable Release Updates Team is being unsubscribed and will not receive messages about this bug report. In the event that you encounter a regression using the package from -updates please report a new bug using ubuntu-bug and tag the bug report regression-update so we can easily find any regresssions.
Yes, i think that would be a wonderfull idea.
Specially because the system does not complain in any way, it just dosnt cache- just spend a few hours on this.
the nfs share was mounted with -o fsc, but that made no difference or error.