Custom Query (684 matches)

The undelying reason os that mondoarchive calls function make_afioballs_and_images_OLD() rather than make_afioballs_and_images() for tape archives. (This happens in function make_those_afios_phase().) At least this explains why there is different behaviour.

Changing the code so it calls make_afioballs_and_images() works fine until towards the end of archiving the normal files where it segfaults. Stupidly, gdb segfaults as well when this happens and I can't even get a backtrace. :-(

The screenout looks like this:

---progress-form---4--- TASK:  [********************]  97% done;  0:32 to go
---progress-form---1--- I am backing up your live filesystem now.
---progress-form---2--- Volume 1: 1,181,312 kilobytes archived so far
---progress-form---3--- Please wait. This may take a couple of hours.
---progress-form---E---
---progress-form---4--- TASK:  [********************]  98% done;  0:21 to go
---progress-form---1--- I am backing up your live filesystem now.
---progress-form---2--- Volume 1: 1,181,312 kilobytes archived so far
---progress-form---3--- Please wait. This may take a couple of hours.
---progress-form---E---
---progress-form---4--- TASK:  [********************]  98% done;  0:21 to go
---progress-form---1--- I am backing up your live filesystem now.
---progress-form---2--- Volume 1: 1,190,016 kilobytes archived so far
---progress-form---3--- Please wait. This may take a couple of hours.
---progress-form---E---
---progress-form---4--- TASK:  [********************]  99% done;  0:10 to go
(Unknown) signal received from OS

Fatal error... Mondoarchive is terminating in response to a signal from the OS
Fatal error... Execution run aborted (main loop)
Execution run ended; result=1
Type 'less /var/log/mondo-archive.log' to see the output log

The final rows of mondo-archive.log are this:

				[Main] libmondo-archive.c->archive_this_fileset#527: command='afio -o -b 131072 -M 16m -Z -P bzip2 -G 3 -T 3k -E /usr/share/mondo/do-not-compress-these /home/mondorescue/tmp/tmp.mondo.30609/tmp.mondo.7027/tmpfs/117.afio.bz2 < /home/mondorescue/tmp/tmp.mondo.30609/tmp.mondo.7027/filelist.117 2>> /var/log/mondo-archive.log'
[Main] libmondo-archive.c->make_afioballs_and_images#1729: AL DEBUG: At end of if in make_afioballs_and_images()
	[Main] libmondo-archive.c->make_afioballs_and_images#1683: Storing set 116
																			[Main] libmondo-files.c->does_file_exist#261: /home/mondorescue/tmp/tmp.mondo.30609/tmp.mondo.7027/filelist.116 exists
																			[Main] libmondo-files.c->does_file_exist#261: /home/mondorescue/tmp/tmp.mondo.30609/tmp.mondo.7027/tmpfs/116.afio.bz2 exists
running: cp -f /home/mondorescue/tmp/tmp.mondo.30609/tmp.mondo.7027/tmpfs/116.afio.bz2 /home/mondorescue/tmp/tmp.mondo.30609/tmp.mondo.7027/tmpfs/backcatalog > /tmp/mondo-run-prog-thing.tmp 2> /tmp/mondo-run-prog-thing.err
--------------------------------start of output-----------------------------
--------------------------------end of output------------------------------
...ran just fine. :-)
	[Main] libmondo-archive.c->make_afioballs_and_images#1699: libmondo-archive.c, make_afioballs_and_images, 1699: Writing EXAT files
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_START_EXTENDED_ATTRIBUTES (fname=xattr_list.116.gz, size=0 K)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_START_EXAT_FILE (fname=xattr_list.116.gz, size=0 K)
	[Main] libmondo-stream.c->write_file_to_stream_from_file#1620: Writing file 'xattr_list.116.gz' to tape (0 KB)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_START_FILE (fname=xattr_list.116.gz, size=0 K)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_STOP_FILE (fname=728eeead8447, size=0 K)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_STOP_EXAT_FILE (fname=xattr_list.116.gz, size=-1 K)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_START_EXAT_FILE (fname=acl_list.116.gz, size=0 K)
	[Main] libmondo-stream.c->write_file_to_stream_from_file#1620: Writing file 'acl_list.116.gz' to tape (0 KB)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_START_FILE (fname=acl_list.116.gz, size=0 K)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_STOP_FILE (fname=76103a4c9048, size=0 K)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_STOP_EXAT_FILE (fname=acl_list.116.gz, size=-1 K)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_STOP_EXTENDED_ATTRIBUTES (fname=xattr_list.116.gz, size=0 K)
																			[Main] libmondo-files.c->does_file_exist#261: /home/mondorescue/tmp/tmp.mondo.30609/tmp.mondo.7027/tmpfs/116.afio.bz2 exists
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_START_AN_AFIO_OR_SLICE (fname=116.afio.bz2, size=13184 K)
	[Main] libmondo-stream.c->write_file_to_stream_from_file#1620: Writing file '116.afio.bz2' to tape (13184 KB)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_START_FILE (fname=116.afio.bz2, size=13184 K)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_STOP_FILE (fname=0d7e3982cbca, size=0 K)
					[Main] libmondo-stream.c->write_header_block_to_stream#1728: BLK_STOP_AN_AFIO_OR_SLICE (fname=finished-writing-file, size=0 K)
[Main] libmondo-archive.c->make_afioballs_and_images#1729: AL DEBUG: At end of if in make_afioballs_and_images()
[Main] libmondo-archive.c->make_afioballs_and_images#1729: AL DEBUG: At end of if in make_afioballs_and_images()
[Main] libmondo-archive.c->make_afioballs_and_images#1729: AL DEBUG: At end of if in make_afioballs_and_images()
			[Main] libmondo-archive.c->create_afio_files_in_background#1222:                                          [8973:0] - archived 117 OK
																			[Main] libmondo-files.c->does_file_exist#258: /home/mondorescue/tmp/tmp.mondo.30609/tmp.mondo.7027/filelist.118 does not exist
		[Main] libmondo-archive.c->create_afio_files_in_background#1233:                                          [8973:0] - exiting
(Unknown) signal received from OS

	[Main] libmondo-fifo.c->kill_buffer#247: kill_buffer() --- command = ps auxww | grep -F "buffer -m 32m -p75 -B -s32768 -o /dev/ht0 2>> /var/log/mondo-archive.log" | grep -Fv grep | awk '{print $2;}' | grep -v PID | tr -s '
' ' ' | awk '{ print $1; }'
	[Main] libmondo-fifo.c->kill_buffer#250: kill_buffer() --- command = kill 2800
running: kill 2800 > /tmp/mondo-run-prog-thing.tmp 2> /tmp/mondo-run-prog-thing.err
--------------------------------start of output-----------------------------
--------------------------------end of output------------------------------
...ran just fine. :-)
[Main] newt-specific.c->fatal_error#377: Fatal error received - 'Mondoarchive is terminating in response to a signal from the OS'
		[Main] newt-specific.c->fatal_error#395: OK, I think I'm the main PID.
	[Main] newt-specific.c->fatal_error#403: I'm going to do some cleaning up now.
			[Main] newt-specific.c->fatal_error#404: killall mindi 2> /dev/null
running: kill `ps auxww | grep " /mondo/do-not " | awk '{print $1;}' | grep -vx "\?"` > /tmp/mondo-run-prog-thing.tmp 2> /tmp/mondo-run-prog-thing.err
--------------------------------start of output-----------------------------
sh: line 0: kill: root: arguments must be process or job IDs
sh: line 0: kill: root: arguments must be process or job IDs
--------------------------------end of output------------------------------
...ran with res=256
running: kill `ps auxww | grep " tmp.mondo " | awk '{print $1;}' | grep -vx "\?"` > /tmp/mondo-run-prog-thing.tmp 2> /tmp/mondo-run-prog-thing.err
--------------------------------start of output-----------------------------
sh: line 0: kill: root: arguments must be process or job IDs
sh: line 0: kill: root: arguments must be process or job IDs
sh: line 0: kill: root: arguments must be process or job IDs
--------------------------------end of output------------------------------
...ran with res=256
running: kill `ps auxww | grep " ntfsclone " | awk '{print $1;}' | grep -vx "\?"` > /tmp/mondo-run-prog-thing.tmp 2> /tmp/mondo-run-prog-thing.err
--------------------------------start of output-----------------------------
sh: line 0: kill: root: arguments must be process or job IDs
sh: line 0: kill: root: arguments must be process or job IDs
sh: line 0: kill: root: arguments must be process or job IDs
--------------------------------end of output------------------------------
...ran with res=256
[Main] libmondo-archive.c->make_afioballs_and_images#1729: AL DEBUG: At end of if in make_afioballs_and_images()
[Main] newt-specific.c->fatal_error#377: Fatal error received - 'Execution run aborted (main loop)'
		[Main] newt-specific.c->fatal_error#395: OK, I think I'm the main PID.
		[Main] newt-specific.c->fatal_error#397: ...I'm already exiting. Give me time, Julian!
	[Main] libmondo-files.c->register_pid#812: Unregistering PID
running: umount /mnt/cdrom > /tmp/mondo-run-prog-thing.tmp 2> /tmp/mondo-run-prog-thing.err
--------------------------------start of output-----------------------------
umount: /mnt/cdrom: not mounted
--------------------------------end of output------------------------------
...ran with res=256
running: rm -Rf /mondo.scratch.* /tmp.mondo.* > /tmp/mondo-run-prog-thing.tmp 2> /tmp/mondo-run-prog-thing.err
--------------------------------start of output-----------------------------
--------------------------------end of output------------------------------
...ran just fine. :-)
running: rm -Rf /home/mondorescue/tmp/tmp.mondo.30609/tmp.mondo.7027 /home/mondorescue/scratch/mondo.scratch.7496/mondo.scratch.11697 > /tmp/mondo-run-prog-thing.tmp 2> /tmp/mondo-run-prog-thing.err
--------------------------------start of output-----------------------------
rm: cannot remove directory `/home/mondorescue/tmp/tmp.mondo.30609/tmp.mondo.7027/tmpfs': Device or resource busy
--------------------------------end of output------------------------------
...ran with res=256
	[Main] libmondo-tools.c->do_libmondo_global_strings_thing#1586: libmondo-tools.c, do_libmondo_global_strings_thing, 1586: Freeing globals

All not very informative I'm afraid.

We can either try to figure out what the differences in behaviour between make_afioballs_and_images_OLD() and than make_afioballs_and_images() are and fix this. Or we can try to fix above segmentation fault.

If you have an idea, Bruno, please let me know. Otherwise I'll keep plodding along...

(As a side note, the kill command fails because it should be $2 in the awk call not $1- that or ps_options needs be different.)

Target: mandriva_2006.0 mandriva_2007.0 fedora_5 rhel_3 rhel_4 suse_10.1 sles_10 debian_3.1 gentoo_1.6

Versions: mondo-2.2.0-2.rhel4 mindi-1.2.0-2.rhel4 On mondoarchive the swap area defined in /etc/fstab as:

LABEL=SWAP-hda2 swap swap defaults 0 0

now gets correctly identified as a swap space (partition).

Test restore/clone: I used boot->mondorestore CD, Automatic, From CD-R and everything ran smoothly to completion of restore with result=0. mondorestore DOES correctly allocate the /dev/hda2 and looking at mondo-restore.log I see that it is doing a 'mkswap /dev/hda2'. The message in the log is: "Setting up swapspace version 1, size = ..."

However, after reboot (single user) the swap space is not active. I do a: swapon -a and it fails because it cannot find the LABEL=SWAP-hda2 device.

I discovered that I needed to do:

mkswap -L 'SWAP-hda2' /dev/hda2

to make the 'swapon -a' work with my current /etc/fstab. (It's the only command I know to label swap space.)

In previous version of mondo/mindi I had been modifying the /etc/fstab to change the swap from a LABEL=... to the standard /dev/hdX (or sdX), and that might be the best choice. However, since there is the attempt by mondo/mindi to deal with Red Hat LABELed swap areas I thought I sould at least share my findings at this version.