Patch applied. Then . . .
amanda@marlin:~/daily$ amcheck -s -otaperscan=taper_lexical daily
Amanda Tape Server Host Check
-----------------------------
WARNING: holding disk '/amanda3': only 104849408 KB free, using nothing
WARNING: Not enough free space specified in amanda.conf
NOTE: Holding disk '/amanda4': 137469952 KB disk space available, using
32612352 KB
NOTE: Holding disk '/amanda5': 104857600 KB disk space available, using
0 KB
amcheck-device: Not a SCALAR reference at
/usr/local/share/perl/5.22.1/Amanda/Taper/Scan/lexical.pm line 102.
Server check took 0.965 seconds
(brought to you by Amanda 3.5.1)
amanda@marlin:~/daily$ ls -lt /tmp/amanda/server/daily/ | head
So, it didn't work (didn't load a tape and check the label), but it didn't crash and there were no
entries in the kern.log.
Following is the end of
/tmp/amanda/server/daily/amcheck-device.20201012151659.debug
Mon Oct 12 15:17:00.503496521 2020: pid 15320: thd-0x1df0800:
amcheck-device: mtx: Storage
Element 39:Full :VolumeTag=000042L7
Mon Oct 12 15:17:00.503538735 2020: pid 15320: thd-0x1df0800:
amcheck-device: mtx: Storage
Element 40:Empty
Mon Oct 12 15:17:00.503577116 2020: pid 15320: thd-0x1df0800:
amcheck-device: mtx: Storage
Element 41:Empty
Mon Oct 12 15:17:00.503615219 2020: pid 15320: thd-0x1df0800:
amcheck-device: mtx: Storage
Element 42:Empty
Mon Oct 12 15:17:00.503657949 2020: pid 15320: thd-0x1df0800:
amcheck-device: mtx: Storage
Element 43:Full :VolumeTag=000043L7
Mon Oct 12 15:17:00.503701295 2020: pid 15320: thd-0x1df0800:
amcheck-device: mtx: Storage
Element 44:Empty
Mon Oct 12 15:17:00.503739869 2020: pid 15320: thd-0x1df0800:
amcheck-device: mtx: Storage
Element 45:Empty
Mon Oct 12 15:17:00.503777939 2020: pid 15320: thd-0x1df0800:
amcheck-device: mtx: Storage
Element 46:Empty
Mon Oct 12 15:17:00.503820736 2020: pid 15320: thd-0x1df0800:
amcheck-device: mtx: Storage
Element 47:Full :VolumeTag=CLN001CU
Mon Oct 12 15:17:00.503867625 2020: pid 15320: thd-0x1df0800:
amcheck-device: mtx: Storage
Element 48:Full :VolumeTag=CLN002CU
Mon Oct 12 15:17:00.504012672 2020: pid 15320: thd-0x1df0800:
amcheck-device: NEO200x48:
updating state
Mon Oct 12 15:17:00.508566057 2020: pid 15320: thd-0x1df0800:
amcheck-device: critical (fatal):
Not a SCALAR reference at
/usr/local/share/perl/5.22.1/Amanda/Taper/Scan/lexical.pm line 102.
/usr/local/lib/amanda/libamanda-3.5.1.so(+0x398c7)[0x7fa8696158c7]
/lib/x86_64-linux-gnu/libglib-2.0.so.0(g_logv+0x1f4)[0x7fa869314a44]
/lib/x86_64-linux-gnu/libglib-2.0.so.0(g_log+0x8f)[0x7fa869314c6f]
/usr/local/share/perl/5.22.1/auto/Amanda/MainLoop/libMainLoop.so(+0x45b7)[0x7fa8659085b7]
/lib/x86_64-linux-gnu/libglib-2.0.so.0(+0x4aab3)[0x7fa86930eab3]
/lib/x86_64-linux-gnu/libglib-2.0.so.0(g_main_context_dispatch+0x15a)[0x7fa86930e04a]
/lib/x86_64-linux-gnu/libglib-2.0.so.0(+0x4a3f0)[0x7fa86930e3f0]
/lib/x86_64-linux-gnu/libglib-2.0.so.0(g_main_context_iteration+0x2c)[0x7fa86930e49c]
/usr/local/lib/amanda/libamanda-3.5.1.so(+0x3b291)[0x7fa869617291]
/usr/local/share/perl/5.22.1/auto/Amanda/MainLoop/libMainLoop.so(_wrap_run_c+0x53)[0x7fa865908d93]
/usr/bin/perl(Perl_pp_entersub+0x48a)[0x4bd5fa]
/usr/bin/perl(Perl_runops_standard+0x16)[0x4b64a6]
/usr/bin/perl(perl_run+0x2f9)[0x443bc9]
/usr/bin/perl(main+0x12b)[0x41cb2b]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf0)[0x7fa869efc840]
/usr/bin/perl(_start+0x29)[0x41cb69]
On 10/11/20 7:04 PM, Nathan Stratton Treadway wrote:
On Sun, Oct 11, 2020 at 14:48:13 -0400, Chris Hoogendyk wrote:
In any case, I currently have both systems flushing to tape using
the traditional taper scan. That may work for now, but it would be
good to track this down. It's also puzzling that it just turned up
R.e. tracking this down: my thought is to see if we can track down the
problem by tweaking the code for lexical.pm -- that way, you can trigger
test runs using "amcheck -otaperscan=taper_lexical" without running the
risk of having a normal cron job attempt to run the code you are in the
middle of modifying.
I've attached a patch which, hopefully, will both fix the
"uninitialized" warning messages that have been appearing in the logs
and also print some debugging info as it loops through the tape inventory
so we can see if it dies in the middle of that loop.
So, when you are ready to investigate further, save a copy of the
original /usr/local/share/perl/5.22.1/Amanda/Taper/Scan/lexical.pm, then
apply the patch to the "live" lexical.pm file, and then try running
amcheck -s (with lexical scan) again.
You may have to fix unblanced quotes or whatever typos in the patched
lines, but hopefully you'll soon get a log file which lists the tapes in
the inventory (but doesn't have the warning lines any more)... at which
point, send me those log lines....
Nathan
----------------------------------------------------------------------------
Nathan Stratton Treadway - [email protected] - Mid-Atlantic region
Ray Ontko & Co. - Software consulting services - http://www.ontko.com/
GPG Key: http://www.ontko.com/~nathanst/gpg_key.txt ID: 1023D/ECFB6239
Key fingerprint = 6AD8 485E 20B9 5C71 231C 0C32 15F3 ADCD ECFB 6239
--
---------------
Chris Hoogendyk
-
O__ ---- Systems Administrator, Retired
c/ /'_ --- Biology & Geosciences Departments
(*) \(*) -- 315 Morrill Science Center III
~~~~~~~~~~ - University of Massachusetts, Amherst
<[email protected]>
---------------
Erdös 4
