[Bug 912334] New: undue start job delay from nofail fstab entries for unattached/unpowered devices
http://bugzilla.opensuse.org/show_bug.cgi?id=912334 Bug ID: 912334 Summary: undue start job delay from nofail fstab entries for unattached/unpowered devices Classification: openSUSE Product: openSUSE Factory Version: 201412* Hardware: Other OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Basesystem Assignee: bnc-team-screening@forge.provo.novell.com Reporter: mrmazda@earthlink.net QA Contact: qa-bugs@suse.de Found By: --- Blocker: --- I have in 13.1 3.12.33, 13.2 3.16.7 and TW 3.18.1 installations on the same machine these identical fstab entries: #/dev/disk/by-label/RS1933-1109 /dev/disk/by-id/ata-ST2000DL003-9VT166_5YD4SNQ6-part1 /disks/esata ntfs-3g noatime,nofail,users,user_id=1050,group_id=1000,fmask=0111,dmask=0000,locale=en_US.UTF-8 0 0 #/dev/disk/by-label/ST32000542ASp1 /dev/disk/by-id/ata-ST2000DL003-9VT166_5YD4SNQ6-part2 /disks/esata2 ext2 noatime,nofail 0 0 #/dev/disk/by-label/ST2000DM001Z2F0RXAS AZBme Seagate /dev/disk/by-id/ata-ST2000DM001-1CH164_Z2F0RXAS-part1 /disks/esata2 ext2 noatime,nofail 0 0 #/dev/disk/by-label/WD20EURSAV-GP1208 AZBme WD /dev/disk/by-id/ata-WDC_WD20EURS-63S48Y0_WD-WCAZAE933621-part1 /disks/esata ext2 noatime,nofail 0 0 Boot proceeds normally on 13.1 and 13.2. On TW there is substantial delay from timing out 4 start jobs for the first 2 of these 4 filesystems when booting without them attached and powered up: # journalctl | grep start | grep disk Jan 08 14:48:12 big41 systemd[1]: Job dev-disk-by\x2did-ata\x2dST2000DL003\x2d9VT166_5YD4SNQ6\x2dpart2.device/start timed out. Jan 08 14:50:38 big41 systemd[1]: Job dev-disk-by\x2did-ata\x2dST2000DL003\x2d9VT166_5YD4SNQ6\x2dpart1.device/start timed out. Jan 08 15:03:00 big41 systemd[1]: Job dev-disk-by\x2did-ata\x2dST2000DL003\x2d9VT166_5YD4SNQ6\x2dpart2.device/start timed out. Jan 08 15:03:00 big41 systemd[1]: Job dev-disk-by\x2did-ata\x2dST2000DL003\x2d9VT166_5YD4SNQ6\x2dpart1.device/start timed out. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=912334 Felix Miata <mrmazda@earthlink.net> changed: What |Removed |Added ---------------------------------------------------------------------------- Flags|needinfo?(mrmazda@earthlink | |.net) | --- Comment #2 from Felix Miata <mrmazda@earthlink.net> --- The problem is it takes an extra 30 seconds (estimated/approximate) to boot TW compared to 13.1 and 13.2 unless these devices are attached and powered up. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=912334 --- Comment #4 from Felix Miata <mrmazda@earthlink.net> --- Noauto means manually mounting once discovery of availability of a device is made by a human. Nofail means it can power up later on and be mounted at that time automatically by the system, big difference, and clear regression in TW. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=912334 --- Comment #6 from Felix Miata <mrmazda@earthlink.net> --- (In reply to Werner Fink from comment #5)
and are you aware that I'll accept this slow down in comparision to a broken dwpendency chain?
In this case of external storage devices switched around because external ports are limited, there isn't any "dependency" involved, other than a mount location I chose, and mounting automatically at connect time. I used to use noauto, but found nofail more convenient, and now you've apparently removed that convenience. Maybe nofail needs a sub-option whether or how long to delay proceeding while waiting. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=912334 http://bugzilla.opensuse.org/show_bug.cgi?id=912334#c20 --- Comment #20 from Felix Miata <mrmazda@earthlink.net> --- Leap has the problem I reported in comment 0. Boot takes more than 2.2 minutes on host gx745 due to two fstab entries for partitions on an unpowered eSATA HD. Booting same machine to 13.1, which has the same two fstab entries, takes 61 seconds. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=912334 http://bugzilla.opensuse.org/show_bug.cgi?id=912334#c24 Franck Bui <fbui@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #24 from Franck Bui <fbui@suse.com> --- Closing. Feel free to re open if you can reproduce on an up to date system. Please note that 13.1 and 13.2 have reached EOL, therefore it doesn't worth reporting against those distros. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=912334 http://bugzilla.opensuse.org/show_bug.cgi?id=912334#c25 --- Comment #25 from Felix Miata <mrmazda@earthlink.net> --- Confirmed comment 0 partitions /dev/disk/by-label/RS1933-1109 and /dev/disk/by-label/ST32000542ASp1 on host big41 work as expected in 42.1, 42.2 and TW. Why this bug was moved from Factory to 13.2 I don't know. I filed it after 13.2 was released; against Factory. It seems to me the only move should have been made to Tumbleweed. -- You are receiving this mail because: You are on the CC list for the bug.
participants (1)
-
bugzilla_noreply＠novell.com