#3047 PR merged: Skip invalid disk drives (zero sized, no media) when saving layout

Labels: enhancement, cleanup, fixed / solved / done

pcahyna opened issue at 2023-09-11 14:18:

Pull Request Details:

  • Type: Enhancement

  • Impact: Normal

  • Reference to related issue (URL): fixes #2958

  • How was this pull request tested?

    • rear savelayout on Fedora 36 with empty USB SD card reader
      I show the relevant log excerpt and the corresponding part of disklayout.conf, if any.

      • original code: No partition label type for 'disk /dev/sdb' (may cause 'rear recover' failure)

        # Disk /dev/sdb
# Format: disk <devname> <size(bytes)> <partition label type>
#disk /dev/sdb 0

      • new code: Ignoring sdb: blockdev: cannot open /dev/sdb: No medium found

        No sdb in disklayout - new code without reverting 0a1d634ed15500bb21f37ac1bbb11c8a4bb11545
        same result - original code only with 0a1d634ed15500bb21f37ac1bbb11c8a4bb11545 reverted
        This reproduces https://github.com/rear/rear/issues/2810 ERROR: Invalid 'disk /dev/sdb' entry (no partition table type for '/dev/sdb') Some latest log messages since the last called script 200_partition_layout.sh: 2023-06-20 19:23:08.249840221 Saving disks and their partitions Some messages from /var/tmp/rear.ZC1Unp4TNvGyfNj/tmp/rear.savelayout.stdout_stderr since the last called script 200_partition_layout.sh: blockdev: cannot open /dev/sdb: No medium found Error: Error opening /dev/sdb: No medium found Use debug mode '-d' for some debug messages or debugscript mode '-D' for full debug messages with 'set -x' output Aborting due to an error, check /home/pcahyna/rear/rear/var/log/rear/rear-pacaziste.log for details Terminated - rear savelayout on Fedora 36 with empty USB SD card reader and AUTOEXCLUDE_DISKS=n - original code: No partition label type for 'disk /dev/sdb' (may cause 'rear recover' failure) /dev/sdb size 0 is not a positive integer ERROR: ==================== BUG in /home/pcahyna/rear/rear/usr/share/rear/layout/save/default/950_verify_disklayout_file.sh line 257: 'Entries in /home/pcahyna/rear/rear/var/lib/rear/layout/disklayout.conf are broken ('rear recover' would fail)' -------------------- Please report it at https://github.com/rear/rear/issues and include all related parts from /home/pcahyna/rear/rear/var/log/rear/rear-pacaziste.log preferably the whole debug information via 'rear -D savelayout' ==================== Some latest log messages since the last called script 950_verify_disklayout_file.sh: 2023-06-20 19:16:28.417175129 Verifying that the entries in /home/pcahyna/rear/rear/var/lib/rear/layout/disklayout.conf are correct 2023-06-20 19:16:28.424548585 Verifying that the 'disk' entries in /home/pcahyna/rear/rear/var/lib/rear/layout/disklayout.conf are correct 2023-06-20 19:16:28.436529921 Verifying that the 'part' entries for /dev/sda in /home/pcahyna/rear/rear/var/lib/rear/layout/disklayout.conf are correct 2023-06-20 19:16:28.446834398 Verifying that the 'part' entries for /dev/sda in /home/pcahyna/rear/rear/var/lib/rear/layout/disklayout.conf specify consecutive partitions 2023-06-20 19:16:28.458051434 Verifying that the 'part' entries for /dev/sdb in /home/pcahyna/rear/rear/var/lib/rear/layout/disklayout.conf are correct 2023-06-20 19:16:28.468396791 Verifying that the 'part' entries for /dev/sdb in /home/pcahyna/rear/rear/var/lib/rear/layout/disklayout.conf specify consecutive partitions 2023-06-20 19:16:28.473731880 Verifying that the 'lvm...' entries in /home/pcahyna/rear/rear/var/lib/rear/layout/disklayout.conf are correct 2023-06-20 19:16:28.506275876 /dev/sdb size 0 is not a positive integer Some messages from /var/tmp/rear.dSRYZZLoczSVake/tmp/rear.savelayout.stdout_stderr since the last called script 950_verify_disklayout_file.sh: 256060514304 1073741824 1048576 254985371648 1074790400 0 Use debug mode '-d' for some debug messages or debugscript mode '-D' for full debug messages with 'set -x' output Aborting due to an error, check /home/pcahyna/rear/rear/var/log/rear/rear-pacaziste.log for details Terminated

        # Disk /dev/sdb
# Format: disk <devname> <size(bytes)> <partition label type>
disk /dev/sdb 0 
# Partitions on /dev/sdb

        This reproduces #2958 - new code: Ignoring sdb: blockdev: cannot open /dev/sdb: No medium found

        No sdb in disklayout - new code without reverting 0a1d634ed15500bb21f37ac1bbb11c8a4bb11545
        same result - full backup and recovery on RHEL 8 with empty USB SD card reader Running 'layout/save' stage ====================== Creating disk layout Ignoring sde: blockdev: cannot open /dev/sde: No medium found Automatically excluding disk /dev/sdb (not used by any mounted filesystem) Marking component '/dev/sdb' as done in /var/lib/rear/layout/disktodo.conf Automatically excluding disk /dev/sdc (not used by any mounted filesystem) Marking component '/dev/sdc' as done in /var/lib/rear/layout/disktodo.conf Automatically excluding disk /dev/sdd (not used by any mounted filesystem) Marking component '/dev/sdd' as done in /var/lib/rear/layout/disktodo.conf Disabling excluded components in /var/lib/rear/layout/disklayout.conf Disabling component 'disk /dev/sdb' in /var/lib/rear/layout/disklayout.conf Disabling component 'disk /dev/sdc' in /var/lib/rear/layout/disklayout.conf Disabling component 'disk /dev/sdd' in /var/lib/rear/layout/disklayout.conf

    • full backup and recovery on RHEL 8 with empty USB SD card reader and AUTOEXCLUDE_DISKS=n Running 'layout/save' stage ====================== Creating disk layout Ignoring sde: blockdev: cannot open /dev/sde: No medium found Disabling excluded components in /var/lib/rear/layout/disklayout.conf

  • Brief description of the changes in this pull request:

    • Introduce function is_disk_valid, which performs checks for disk usability beyond validating the device name, and use it when saving the disk layout. In some cases the device name may be valid, but there are no data, typically because it is a drive with removable media and there is no medium in the tray. Happens typically with card (e.g. SD card) readers with empty slot.
      This is a normal occurrence, so do not Error out, only display a message and skip the device.
    • Revert commit 0a1d634ed15500bb21f37ac1bbb11c8a4bb11545 . We now skip disks with no data (like when there is no medium), so incomplete disk entries (without partition type) should not occur anymore. Restore the code that aborted when such disks were encountered.
      Incomplete entries should not be allowed to occur, as they could confuse the layout restoration code. Moreover, the layout restoration wipes all disks in the layout, so if during layout restoration there happens to be a medium in the drive that was empty during layout save, the data on the medium would get overwritten and lost. And if there is not medium, the layout recreation script would fail.
      See the discussion at https://github.com/rear/rear/issues/2958#issuecomment-1479588829

jsmeix commented at 2023-09-12 06:51:

@pcahyna
thank you for this enhancement and
the cleanup of the old clumsy code
and for your explanatory description
of the various cases (in particular
things like the special DASD case)!

pcahyna commented at 2023-09-12 15:13:

@jsmeix I believe I addressed all review comments in last 3 commits.

jsmeix commented at 2023-09-13 05:56:

@pcahyna
thank you for the fine tuning.
All looks perfect to me.
Feel free to merge it as you like.

[Export of Github issue for rear/rear.]