GNOME Bugzilla – Bug 425129
crash in icaltimezone.c: zone->tznames == NULL
Last modified: 2007-12-04 15:32:42 UTC
What were you doing when the application crashed? Evolution simply doesn't launch, when I start it. Distribution: Ubuntu 6.10 (edgy) Gnome Release: 2.16.1 2006-10-02 (Ubuntu) BugBuddy Version: 2.16.0 Memory status: size: 222601216 vsize: 222601216 resident: 25882624 share: 15179776 rss: 25882624 rss_rlim: -1 CPU usage: start_time: 1175428734 rtime: 94 utime: 88 stime: 6 cutime:0 cstime: 0 timeout: 0 it_real_value: 0 frequency: 100 Backtrace was generated from '/usr/bin/evolution-2.8' (no debugging symbols found) Using host libthread_db library "/lib/libthread_db.so.1". (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread 47161910119792 (LWP 8077)] (no debugging symbols found) 0x00002ae4b89a5595 in waitpid () from /lib/libc.so.6
+ Trace 124037
Thread 1 (Thread 47161910119792 (LWP 8077))
Thanks for taking the time to report this bug. Unfortunately, that stack trace is missing some elements that will help a lot to solve the problem, so it will be hard for the developers to fix that crash. Can you get us a stack trace with debugging symbols? Please see http://live.gnome.org/GettingTraces for more information on how to do so. Thanks in advance!
Had a look on the link above... The problem is that my evolution version is 2.8.1-0ubuntu4.1 and there is only -dbgsym for 2.8.1-0ubuntu4. I can't install it from repository because it doesn't like that my version is higher... What should I do?
There does appear to be dbgsyms for 2.8.1-0ubuntu4.1 in the repository. Not sure why they're not appear to you. You could try the folder http://people.ubuntu.com/~pitti/ddebs/pool/main/e/evolution/ and installing the .ddeb's directly? ("sudo dpkg -i something.ddeb" at the command line)
Downloaded and installed evolution and plugins -dbgsym from given page. I have started evolution once again, but Bug Buddy generated almost the same report... So, I tried gdb and the comands that were listed on "GettingTraces/Details" page. Here what we have:
+ Trace 124581
Thread 1 (Thread 47263941999984 (LWP 13841))
Is it enough of debug info? Or should I do anything else?
*** Bug 421685 has been marked as a duplicate of this bug. ***
*** Bug 421911 has been marked as a duplicate of this bug. ***
*** Bug 419675 has been marked as a duplicate of this bug. ***
(In reply to comment #4) > Downloaded and installed evolution and plugins -dbgsym from given page. I have > started evolution once again, but Bug Buddy generated almost the same report... > So, I tried gdb and the comands that were listed on "GettingTraces/Details" > page. Here what we have: Thanks, that's a much better better stacktrace. However, we're still missing information from libecal. You should be able to install the libecal1.2-7-dbgsym package (available at http://people.ubuntu.com/~pitti/ddebs/pool/main/e/evolution-data-server/ if you can't find it), and then we can get the complete trace. Thanks for all your help!
Debug backtrace on https://bugs.launchpad.net/ubuntu/+source/evolution/+bug/99878:
+ Trace 124595
Marking as NEW now we've got duplicates and a detailed stacktrace.
I've installed libecall and ran gdb as did before (gdb evolution -> run -> thread apply all bt), but report is similar to previous one:
+ Trace 124626
Thread 1 (Thread 47740351135088 (LWP 18595))
Am I doing anything wrong?
(In reply to comment #11) > I've installed libecall and ran gdb as did before (gdb evolution -> run -> > thread apply all bt), but report is similar to previous one: It's similar, but now frame #1 has the proper debugging symbols, so you've done the right thing. We now have enough information to hand over this bug to the main Evolution developers (hence why this bug has gone from being UNCONFIRMED/NEEDINFO to NEW). Thanks for your time!
You are welcome! Glad to be helpfull :)
*** Bug 425908 has been marked as a duplicate of this bug. ***
*** Bug 426181 has been marked as a duplicate of this bug. ***
2 zones don't have TZNAME set : zoneinfo/Australia/Perth.ics zoneinfo/Asia/Jerusalem.ics this leads zone->tznames to be NULL
*** Bug 429009 has been marked as a duplicate of this bug. ***
*** Bug 426144 has been marked as a duplicate of this bug. ***
*** Bug 428926 has been marked as a duplicate of this bug. ***
*** Bug 429225 has been marked as a duplicate of this bug. ***
Created attachment 86289 [details] [review] Don't crash when a zone does not have TZNAME
Created attachment 86290 [details] [review] Add a timezone for Perth Here is a patch for Perth but as the ics contains : PRODID:-//Ximian//NONSGML Evolution Olson-VTIMEZONE Converter//EN I guess it is generated from elsewhere where it should be fixed...
*** Bug 430211 has been marked as a duplicate of this bug. ***
Created attachment 86997 [details] [review] resolve timezone check prob;em With new updated timezones zone->tznames needs check for NULL
This is exactly the patch that I add already attached :)
Marking patch from comment 24 as obsolete due to comment 25.
*** Bug 430378 has been marked as a duplicate of this bug. ***
*** Bug 429984 has been marked as a duplicate of this bug. ***
*** Bug 432074 has been marked as a duplicate of this bug. ***
*** Bug 432077 has been marked as a duplicate of this bug. ***
*** Bug 432581 has been marked as a duplicate of this bug. ***
*** Bug 429410 has been marked as a duplicate of this bug. ***
*** Bug 433705 has been marked as a duplicate of this bug. ***
*** Bug 435554 has been marked as a duplicate of this bug. ***
*** Bug 436906 has been marked as a duplicate of this bug. ***
*** Bug 438108 has been marked as a duplicate of this bug. ***
*** Bug 438687 has been marked as a duplicate of this bug. ***
*** Bug 437474 has been marked as a duplicate of this bug. ***
*** Bug 438310 has been marked as a duplicate of this bug. ***
*** Bug 439411 has been marked as a duplicate of this bug. ***
bug 435218 is probably also a dup
*** Bug 438848 has been marked as a duplicate of this bug. ***
*** Bug 439551 has been marked as a duplicate of this bug. ***
*** Bug 435218 has been marked as a duplicate of this bug. ***
*** Bug 440022 has been marked as a duplicate of this bug. ***
*** Bug 440133 has been marked as a duplicate of this bug. ***
The patch avoiding the crash is _really_ trivial, can't any evolution maintainer have a look at it, or do we need more duplicates ?
*** Bug 440567 has been marked as a duplicate of this bug. ***
*** Bug 440912 has been marked as a duplicate of this bug. ***
It would be safe to add a null check for the tzname string as well. Please commit the patches after incorporating the above mentioned change to svn HEAD and stable branch.
I committed the patches (with an additional check for tzname) zoneinfo/Asia/Jerusalem.ics still has no TZNAME but I could not find it
*** Bug 442240 has been marked as a duplicate of this bug. ***
(In reply to comment #51) > I committed the patches (with an additional check for tzname) > > zoneinfo/Asia/Jerusalem.ics still has no TZNAME but I could not find it > The Tznames for standard and daylight zones for Jerusalem are IST and IDT respectively. You can update them as well. Timezone files are perfectly valid if they do not have the tzname as well. Thanks for the patches !!! IMHO you can now mark the bug as resolved.
Thank you for the info ! The file is valid without tzname but it will be ignored if we look for the timezone by name like in this function so I prefer adding it. Closing the bug.
Sure. Please add the tzname information.
*** Bug 442713 has been marked as a duplicate of this bug. ***
*** Bug 442923 has been marked as a duplicate of this bug. ***
*** Bug 443046 has been marked as a duplicate of this bug. ***
*** Bug 443273 has been marked as a duplicate of this bug. ***
*** Bug 443154 has been marked as a duplicate of this bug. ***
*** Bug 443245 has been marked as a duplicate of this bug. ***
*** Bug 443260 has been marked as a duplicate of this bug. ***
*** Bug 443261 has been marked as a duplicate of this bug. ***
*** Bug 443264 has been marked as a duplicate of this bug. ***
*** Bug 443495 has been marked as a duplicate of this bug. ***
*** Bug 443508 has been marked as a duplicate of this bug. ***
*** Bug 443551 has been marked as a duplicate of this bug. ***
*** Bug 443544 has been marked as a duplicate of this bug. ***
*** Bug 443543 has been marked as a duplicate of this bug. ***
*** Bug 443621 has been marked as a duplicate of this bug. ***
*** Bug 443690 has been marked as a duplicate of this bug. ***
*** Bug 444424 has been marked as a duplicate of this bug. ***
*** Bug 445755 has been marked as a duplicate of this bug. ***
*** Bug 445743 has been marked as a duplicate of this bug. ***
*** Bug 445428 has been marked as a duplicate of this bug. ***
*** Bug 445554 has been marked as a duplicate of this bug. ***
*** Bug 444807 has been marked as a duplicate of this bug. ***
from bug 444807: >> Unfortunately, order of expression evaluation is unspecified by standards in C language. I afraid, patch http://bugzilla.gnome.org/attachment.cgi?id=86289&action=view : if (z_offset == offset && zone->tznames && !strcmp(tzname, zone->tznames)) may not completely resolve a crash problem, because in something combinations of C-compilers, OS-es and optimization options may rearrange expressions evaluation and, as a consequence, strcmp(tzname, zone->tznames) may be evaluated before zone->tznames. In my opinion, this code may be more safe expression: if (zone->tznames == NULL) continue; <<
*** Bug 445670 has been marked as a duplicate of this bug. ***
*** Bug 445817 has been marked as a duplicate of this bug. ***
(In reply to comment #78) > from bug 444807: > > Unfortunately, order of expression evaluation is unspecified by standards in C > language That's wrong && order is specified : all logical operators are evaluated left-to-right, stopping when the result is known (when false is met for && and when true is met for ||). What is not specified is the order of evaluation of function arguments or arithmetic expressions.
*** Bug 446641 has been marked as a duplicate of this bug. ***
*** Bug 446666 has been marked as a duplicate of this bug. ***
*** Bug 446678 has been marked as a duplicate of this bug. ***
*** Bug 446762 has been marked as a duplicate of this bug. ***
*** Bug 446760 has been marked as a duplicate of this bug. ***
(In reply to comment #81) > (In reply to comment #78) > > from bug 444807: > > > > Unfortunately, order of expression evaluation is unspecified by standards in C > > language > > That's wrong && order is specified : all logical operators are evaluated > left-to-right, stopping when the result is known (when false is met for && and > when true is met for ||). > > What is not specified is the order of evaluation of function arguments or > arithmetic expressions. > Unfortunately, I don't have ANSI C standard near at hand now, but I think -- you are right & I am wrong ;). Thanks for good comments, Pascal!
*** Bug 446952 has been marked as a duplicate of this bug. ***
*** Bug 447200 has been marked as a duplicate of this bug. ***
*** Bug 447409 has been marked as a duplicate of this bug. ***
*** Bug 447523 has been marked as a duplicate of this bug. ***
*** Bug 447531 has been marked as a duplicate of this bug. ***
*** Bug 447791 has been marked as a duplicate of this bug. ***
*** Bug 447981 has been marked as a duplicate of this bug. ***
*** Bug 447913 has been marked as a duplicate of this bug. ***
*** Bug 448320 has been marked as a duplicate of this bug. ***
*** Bug 449106 has been marked as a duplicate of this bug. ***
*** Bug 449421 has been marked as a duplicate of this bug. ***
*** Bug 449518 has been marked as a duplicate of this bug. ***
*** Bug 449560 has been marked as a duplicate of this bug. ***
*** Bug 449703 has been marked as a duplicate of this bug. ***
*** Bug 449753 has been marked as a duplicate of this bug. ***
*** Bug 449946 has been marked as a duplicate of this bug. ***
*** Bug 450625 has been marked as a duplicate of this bug. ***
*** Bug 450759 has been marked as a duplicate of this bug. ***
*** Bug 450553 has been marked as a duplicate of this bug. ***
*** Bug 451214 has been marked as a duplicate of this bug. ***
*** Bug 451279 has been marked as a duplicate of this bug. ***
*** Bug 451477 has been marked as a duplicate of this bug. ***
*** Bug 451548 has been marked as a duplicate of this bug. ***
*** Bug 451561 has been marked as a duplicate of this bug. ***
*** Bug 451602 has been marked as a duplicate of this bug. ***
*** Bug 451771 has been marked as a duplicate of this bug. ***
*** Bug 451942 has been marked as a duplicate of this bug. ***
*** Bug 452190 has been marked as a duplicate of this bug. ***
*** Bug 450855 has been marked as a duplicate of this bug. ***
*** Bug 452459 has been marked as a duplicate of this bug. ***
*** Bug 452243 has been marked as a duplicate of this bug. ***
*** Bug 450898 has been marked as a duplicate of this bug. ***
*** Bug 452898 has been marked as a duplicate of this bug. ***
*** Bug 452931 has been marked as a duplicate of this bug. ***
*** Bug 453040 has been marked as a duplicate of this bug. ***
*** Bug 453019 has been marked as a duplicate of this bug. ***
*** Bug 453351 has been marked as a duplicate of this bug. ***
*** Bug 453992 has been marked as a duplicate of this bug. ***
*** Bug 454426 has been marked as a duplicate of this bug. ***
*** Bug 454550 has been marked as a duplicate of this bug. ***
*** Bug 454670 has been marked as a duplicate of this bug. ***
*** Bug 455988 has been marked as a duplicate of this bug. ***
*** Bug 455524 has been marked as a duplicate of this bug. ***
*** Bug 455432 has been marked as a duplicate of this bug. ***
*** Bug 455431 has been marked as a duplicate of this bug. ***
*** Bug 455420 has been marked as a duplicate of this bug. ***
*** Bug 455161 has been marked as a duplicate of this bug. ***
*** Bug 455156 has been marked as a duplicate of this bug. ***
*** Bug 454683 has been marked as a duplicate of this bug. ***
*** Bug 453936 has been marked as a duplicate of this bug. ***
*** Bug 453887 has been marked as a duplicate of this bug. ***
*** Bug 453793 has been marked as a duplicate of this bug. ***
*** Bug 453709 has been marked as a duplicate of this bug. ***
*** Bug 453691 has been marked as a duplicate of this bug. ***
*** Bug 453682 has been marked as a duplicate of this bug. ***
*** Bug 456171 has been marked as a duplicate of this bug. ***
*** Bug 456162 has been marked as a duplicate of this bug. ***
The fix for this crash was committed on 2007-05-24: http://svn.gnome.org/viewcvs/libical/branches/gnome-2-18/src/libical/icaltimezone.c?r1=591&r2=602 This fix is included in evolution-data-server version 1.10.2. If you run Fedora 7, an updated version of evolution-data-server (containing the fix for this issue) has been available since 2007-06-16.
*** Bug 461595 has been marked as a duplicate of this bug. ***
*** Bug 456210 has been marked as a duplicate of this bug. ***
*** Bug 466137 has been marked as a duplicate of this bug. ***
*** Bug 467378 has been marked as a duplicate of this bug. ***
*** Bug 469514 has been marked as a duplicate of this bug. ***
*** Bug 471140 has been marked as a duplicate of this bug. ***
*** Bug 473258 has been marked as a duplicate of this bug. ***
*** Bug 475310 has been marked as a duplicate of this bug. ***
*** Bug 475396 has been marked as a duplicate of this bug. ***