Summary: | sys-fs/lvm2-2.02.97-r1 - Errors on LV create/delete operations in systemd-196 environment. | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Pavel Volkov <ao> |
Component: | [OLD] Core system | Assignee: | Robin Johnson <robbat2> |
Status: | RESOLVED DUPLICATE | ||
Severity: | normal | CC: | agk, alexander, cardoe, lists, nikoli |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Pavel Volkov
2013-01-11 20:16:50 UTC
"0.11.8" in my previous post related to openrc. Looks very similar to an old kernel bug. See this links: https://bugzilla.redhat.com/show_bug.cgi?id=727925 https://code.google.com/p/chromium/issues/detail?id=92762 https://lkml.org/lkml/2011/8/18/112 Paul, can you test this with kernel 3.6.x? I downgraded to gentoo-sources 3.6.11 and still had the same error. I currently run kernel 3.7.3. A little more verbose output: # lvcreate -vv -L5G -nmio xes Setting activation/monitoring to 1 Setting global/locking_type to 1 Setting global/wait_for_locks to 1 File-based locking selected. Setting global/locking_dir to /run/lock/lvm Setting global/prioritise_write_locks to 1 Getting target version for linear Found linear target v1.1.0. Getting target version for striped Found striped target v1.5.0. Setting logging type to disk Setting activation/mirror_region_size to 512 Finding volume group "xes" Locking /run/lock/lvm/V_xes WB /dev/sda: size is 125045424 sectors /dev/sda1: size is 67584 sectors /dev/sda1: size is 67584 sectors /dev/sda1: No label detected /dev/sda2: size is 346112 sectors /dev/sda2: size is 346112 sectors /dev/sda2: No label detected /dev/sda3: size is 31457280 sectors /dev/sda3: size is 31457280 sectors /dev/sda3: No label detected /dev/sda4: size is 93171712 sectors /dev/sda4: size is 93171712 sectors /dev/sda4: lvm2 label detected at sector 1 /dev/sdb: size is 5860533168 sectors /dev/sdb1: size is 5860530176 sectors /dev/sdb1: size is 5860530176 sectors /dev/sdb1: lvm2 label detected at sector 1 Archiving volume group "xes" metadata (seqno 1). Creating logical volume mio Extending segment type, striped Setting allocation/mirror_logs_require_separate_pvs to 0 Setting allocation/maximise_cling to 1 Creating volume group backup "/etc/lvm/backup/xes" (seqno 2). Locking LV OyBR38yg9Xd8yeL4S3ED2sLeTgOA7OdatjP7EbDAZC1Wk8FL2m8UCVwq1SuJNUlJ (R) Finding volume group for uuid OyBR38yg9Xd8yeL4S3ED2sLeTgOA7OdatjP7EbDAZC1Wk8FL2m8UCVwq1SuJNUlJ /dev/sda4: lvm2 label detected at sector 1 Found volume group "xes" activation/volume_list configuration setting not defined: Checking only host tags for xes/mio /dev/sda4: read_ahead is 256 sectors Setting activation/use_mlockall to 0 Locking memory Creating xes-mio Loading xes-mio table (253:5) Resuming xes-mio (253:5) Unlocking memory Clearing start of logical volume "mio" /dev/xes/mio: stat failed: No such file or directory /dev/xes/mio: not found: device not cleared Aborting. Failed to wipe start of new LV. Locking LV OyBR38yg9Xd8yeL4S3ED2sLeTgOA7OdatjP7EbDAZC1Wk8FL2m8UCVwq1SuJNUlJ (NL) Finding volume group for uuid OyBR38yg9Xd8yeL4S3ED2sLeTgOA7OdatjP7EbDAZC1Wk8FL2m8UCVwq1SuJNUlJ Found volume group "xes" Setting activation/use_mlockall to 0 Locking memory Removing xes-mio (253:5) Setting devices/issue_discards to 0 Unlocking memory Creating volume group backup "/etc/lvm/backup/xes" (seqno 3). semid 491521: semop failed for cookie 0xd4dd624: incorrect semaphore state Failed to set a proper state for notification semaphore identified by cookie value 223204900 (0xd4dd624) to initialize waiting for incoming notifications. Unlocking /run/lock/lvm/V_xes I have recently migrated to systemd and can confirm this issue. Same error with systemd-197-r1, udev-197-r4 and lvm2-2.02.97-r1. (In reply to comment #4) > I have recently migrated to systemd and can confirm this issue. Same error > with systemd-197-r1, udev-197-r4 and lvm2-2.02.97-r1. and hardened-sources-3.7.4-r1 Hit the same errors with * gentoo-sources-3.8.1 * systemd-198 (and 197.x before as well) * lvm2-2.02.98 * udev-198 ~am64 Can't create new LVs oder snapshots right now. Example: # lvcreate --size 1000M --name test VG03 -v Setting logging type to disk Finding volume group "VG03" Archiving volume group "VG03" metadata (seqno 1104). Creating logical volume test Creating volume group backup "/etc/lvm/backup/VG03" (seqno 1105). Found volume group "VG03" activation/volume_list configuration setting not defined: Checking only host tags for VG03/test Creating VG03-test Loading VG03-test table (253:8) Resuming VG03-test (253:8) Clearing start of logical volume "test" /dev/VG03/test: not found: device not cleared Aborting. Failed to wipe start of new LV. Found volume group "VG03" Removing VG03-test (253:8) device-mapper: remove ioctl on failed: Das Gerät oder die Ressource ist belegt Creating volume group backup "/etc/lvm/backup/VG03" (seqno 1106). semid 720896: semop failed for cookie 0xd4d1df0: incorrect semaphore state Failed to set a proper state for notification semaphore identified by cookie value 223157744 (0xd4d1df0) to initialize waiting for incoming notifications. This is related to bug #453594 comment #19 (In reply to comment #7) > This is related to bug #453594 comment #19 I have this issue with dracut, so this is not related. OK. This setting in kernel config helped me: CONFIG_UEVENT_HELPER_PATH="" https://www.redhat.com/archives/linux-lvm/2012-August/msg00025.html PS. Creating snapshots still fails with lvmetad, but this is another story. I confirm CONFIG_UEVENT_HELPER_PATH method works. Maybe kernel options test should be added into the ebuild? I proposed a patch here: https://bugs.gentoo.org/show_bug.cgi?id=453594#c26 *** This bug has been marked as a duplicate of bug 453594 *** |