Bug#927892: grub-ieee1275-bin: Please add bootinfo.txt for sparc and sparc64
On 24/04/2019 21:30, Frank Scheiner wrote: > On 4/24/19 22:07, John Paul Adrian Glaubitz wrote: >> On 4/24/19 9:46 PM, Mark Cave-Ayland wrote: >>> For reference the CHRP bootinfo.txt isn't a configuration file, but is >>> actually >>> parsed by CHRP-compliant open firmwares directly - SPARC firmwares, >>> including >>> OpenBIOS don't support them. Can you explain why grub on SPARC is trying to >>> access >>> this file? >> >> Good point, I didn't notice that there is a CHRP header and assumed that we >> need >> this file on sparc and sparc64 as well. >> >> I haven't actually done any runtime tests yet, I replaced the configuration >> files >> from ppc64el for grub-ieee1275 and adapted them for sparc64. debian-installer >> builds fine with those changes, but as I said, I haven't tested anything yet. >> >> I am very welcome for any pointers though for what else is needed to use GRUB >> for CD boot on sparc64. > > `grub-mkimage` has a (target) format named "sparc64-ieee1275-cdcore". > But I'm unsure if this is of any use for us, or needed at all, because > if GRUB can load kernel and initramfs from disc, it should also be able > to load its modules from disc, so a small GRUB image and separate > modules (like for on-disk or diskless installations) could also work. > > OTOH: if a "sparc64-ieee1275-cdcore" image and a corresponding grub.cfg > works for disc booting, we could just exchange the SILO image and config > for these. Is it possible to see the source for sparc64-ieee1275-cdcore anywhere? A quick search doesn't return anything obvious. And from memory grub is working when installed onto the disk? If so, where is the source of the disk image for comparison? ATB, Mark.
Bug#927892: grub-ieee1275-bin: Please add bootinfo.txt for sparc and sparc64
On 24/04/2019 18:26, John Paul Adrian Glaubitz wrote: > Source: grub2 > Version: 2.02+dfsg1-17 > Severity: normal > User: debian-sp...@lists.debian.org > Usertags: sparc64 > > Hello! > > I have started working on switching the bootloader for debian-installer from > silo to grub-ieee1275 on sparc and sparc64. > > For that, I have looked at the corresponding code in debian-installer for > ppc64el and noticed that the configuration file build/config/ppc64el.cfg > requires a "bootinfo.txt" configuration file which is part of the > grub-ieee1275-bin package. > > Thus, in order to switch to grub-ieee1275 on sparc and sparc64, we need > to add the bootinfo.txt there as well. Could you install the bootinfo.txt > file for sparc and sparc64 as well? > > Plus, it might be a good idea to adjust the GRUB version number as well > as the name of the boot script to be a generic name like debian-installer.elf > and not powerpc.elf. > > Currently, the header of bootinfo.txt looks like this: > > > grub 2.02~beta3 > grub 2.02~beta3 > boot :\boot\grub\powerpc.elf > > > Thanks, > Adrian Hi Adrian, For reference the CHRP bootinfo.txt isn't a configuration file, but is actually parsed by CHRP-compliant open firmwares directly - SPARC firmwares, including OpenBIOS don't support them. Can you explain why grub on SPARC is trying to access this file? ATB, Mark.
Bug#841044: Problem with gem network card
On 27/08/18 10:39, Mathieu Malaterre wrote: > Christian, > > On Mon, Oct 17, 2016 at 9:32 AM Mathieu Malaterre wrote: >> >> Hi, >> >> On Fri, Oct 14, 2016 at 4:45 PM, Thadeu Lima de Souza Cascardo >> wrote: >>> There is certainly a way to automate this, but my question on whether >>> that sequence would fix the problem was intended to more easily fix a >>> real bug in the driver. >>> >>> I can try to help fix the driver on my free time as a volunteer. >>> Christian, would you be willing to test patches to the driver? That >>> would require some back and forth between us and some patience of yours, >>> as I would be doing it on my spare time, considering I have family and >>> other projects I would like to attend to as well. >>> >>> I don't have a PPC Macbook of my own, but I used to fix Linux drivers on >>> PPC64 for a living not too long ago. >> >> That's a good news ! Feel free to use the bug report #841044 to send >> your patch (and/or updates). > > Any update on the patch ? > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=841044#30 FWIW QEMU now has sungem emulation which may help when trying to track down overflow errors such as this. ATB, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 25/02/18 00:16, Samuel Thibault wrote: I have investigated, and yes there are leaks there. It'll be quite invasive to fix, so I won't be able to push that to stretch. But I believe if you can confirm that at least the crashes are gone, we can push the fix to stretch. Hi Samuel, Just to confirm that after 2 days I've seen no console warnings/errors or experienced any crashes here with your proposed stretch packages, so I believe they are working well. I've now just upgraded to your latest -20 packages to give them a more thorough testing, and will keep an eye on the memory usage over the course of the day. Many thanks, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 25/02/18 00:16, Samuel Thibault wrote: Mark Cave-Ayland, on sam. 24 févr. 2018 12:53:58 +, wrote: Not that this was more than a casual observation but I was pondering the possibility of a memory leak somewhere in the C code? I have investigated, and yes there are leaks there. It'll be quite invasive to fix, so I won't be able to push that to stretch. But I believe if you can confirm that at least the crashes are gone, we can push the fix to stretch. Understood. I have another couple of days development ahead so I will keep with the stretch packages for another couple of days just to be sure, then I'll switch over to your latest sid packages and give those a good testing. Also thanks once again for your work on fixing up the memory leaks - I had a peek at the git repository and I can see why you consider the change to be quite invasive... ATB, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 23/02/18 23:22, Samuel Thibault wrote: Hello, Mark Cave-Ayland, on mar. 20 févr. 2018 10:55:43 +, wrote: Done. I've just installed the 0.33.3-13+deb9u1~0_amd64 packages from the above URL (including debug symbols, just in case) and I'll report back in a few days if everything is still looking good. So, how is it going? Hi Samuel, Unfortunately I've been ill for most of this week, however I managed to run netbeans for a day with no crashes and no console warnings using these packages which is brilliant! One thing I did notice was that by the end of that day, netbeans was running quite slow (according to the netbeans widget JVM usage was ~0.5G) however the process in top was showing over 1.5G memory usage. Not that this was more than a casual observation but I was pondering the possibility of a memory leak somewhere in the C code? I hope to be back at work full time next week so I can do more testing then to compare usage with and without accessibility features enabled. ATB, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 20/02/18 10:36, Samuel Thibault wrote: Control: reassign -1 java-atk-wrapper Mark Cave-Ayland, on mar. 20 févr. 2018 10:24:02 +, wrote: I've now installed the -17 packages and after a few minutes developing with lots of autocomplete, I can confirm that the GLib warning/critical messages no longer appear in the console. Cool :) With the standard stretch packages I would get a full netbeans crash maybe 4-5 times a day, so while I'm not yet confident enough to say that this bug has been fixed, this is looking promising so far. Could you also try the stretch update I have put on deb http://people.debian.org/~sthibault/tmp/stretch-tmp ./ so I could upload it to Stretch to fix the issue there for everybody too. Done. I've just installed the 0.33.3-13+deb9u1~0_amd64 packages from the above URL (including debug symbols, just in case) and I'll report back in a few days if everything is still looking good. ATB, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 20/02/18 09:28, Samuel Thibault wrote: Mark Cave-Ayland, on mar. 20 févr. 2018 09:00:14 +, wrote: Fantastic news! I've just had a look at pool and it looks like quite a few of the archs have been built, but amd64 isn't there yet. I'll check again a bit later and install them when they appear. Note that you can already fetch directly from incoming with deb http://incoming.debian.org/debian-buildd buildd-unstable main contrib non-free Are upstream still maintaining java-atk-wrapper? I'm not sure :/ I've now installed the -17 packages and after a few minutes developing with lots of autocomplete, I can confirm that the GLib warning/critical messages no longer appear in the console. With the standard stretch packages I would get a full netbeans crash maybe 4-5 times a day, so while I'm not yet confident enough to say that this bug has been fixed, this is looking promising so far. Thank you! ATB, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 19/02/18 23:39, Samuel Thibault wrote: Hello, Mark Cave-Ayland, on lun. 19 févr. 2018 21:05:00 +, wrote: On 19/02/18 09:03, Samuel Thibault wrote: Mark Cave-Ayland, on lun. 19 févr. 2018 09:00:36 +, wrote: Are the debug packages available anywhere at http://debug.mirrors.debian.org/? If so, I can install them and dig a bit more. It's all on http://debug.mirrors.debian.org/debian-debug/pool/main/j/java-atk-wrapper/ You can also use deb http://debug.mirrors.debian.org/debian-debug/ unstable-debug main in your sources.list Okay, sorry about the delay - attached is the output from the -16 packages when attempting to autocomplete with CTRL-space in netbeans, Ok, these are other issues, which at some point I managed to reproduce with ctrl-space indeed. I have uploaded -17 with fixes against them, to be available soon. Fantastic news! I've just had a look at pool and it looks like quite a few of the archs have been built, but amd64 isn't there yet. I'll check again a bit later and install them when they appear. BTW I had a look at the debian/patches list and there are quite a few patches with upstream bug reports which don't have a response. Are upstream still maintaining java-atk-wrapper? Or will you be able to drop more of the debian-specific patches next time you merge from upstream? ATB, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 19/02/18 09:03, Samuel Thibault wrote: Mark Cave-Ayland, on lun. 19 févr. 2018 09:00:36 +, wrote: Are the debug packages available anywhere at http://debug.mirrors.debian.org/? If so, I can install them and dig a bit more. It's all on http://debug.mirrors.debian.org/debian-debug/pool/main/j/java-atk-wrapper/ You can also use deb http://debug.mirrors.debian.org/debian-debug/ unstable-debug main in your sources.list Okay, sorry about the delay - attached is the output from the -16 packages when attempting to autocomplete with CTRL-space in netbeans, using the same gdb setup as before to catch the CRITICAL/WARNING messages emitted on the console: (gdb) break g_log if log_level == G_LOG_LEVEL_CRITICAL (gdb) break g_log if log_level == G_LOG_LEVEL_WARNING (gdb) handle SIGSEGV nostop noprint pass ATB, Mark. libatk-wrapper-java-jni_0.33.3-16-gdb-debug.txt.xz Description: application/xz
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 19/02/18 08:39, Samuel Thibault wrote: Thanks! The corresponding backtrace look like these warnings are probably fixed by the -16 upload too. Samuel Okay I've just installed the -16 packages locally and I'm still seeing warnings on the console when hitting CTRL-space during autocomplete: (java:8938): GLib-CRITICAL **: g_hash_table_iter_init: assertion 'iter != NULL' failed (java:8938): GLib-CRITICAL **: g_hash_table_iter_next: assertion 'iter != NULL' failed (java:8938): GLib-CRITICAL **: g_hash_table_iter_init: assertion 'iter != NULL' failed (java:8938): GLib-CRITICAL **: g_hash_table_iter_next: assertion 'iter != NULL' failed (java:8938): GLib-GObject-WARNING **: invalid uninstantiatable type '' in cast to 'AtkObject' (java:8938): GLib-GObject-WARNING **: instance of invalid non-instantiatable type '' (java:8938): GLib-GObject-CRITICAL **: g_signal_emit_by_name: assertion 'G_TYPE_CHECK_INSTANCE (instance)' failed (java:8938): GLib-GObject-WARNING **: invalid uninstantiatable type '' in cast to 'GObject' (java:8938): GLib-GObject-WARNING **: invalid uninstantiatable type '' in cast to 'GObject' (java:8938): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed (java:8938): GLib-GObject-WARNING **: invalid uninstantiatable type '' in cast to 'GObject' (java:8938): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed (java:8938): GLib-GObject-WARNING **: invalid uninstantiatable type '' in cast to 'GObject' (java:8938): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed (java:8938): GLib-GObject-WARNING **: invalid uninstantiatable type '' in cast to 'GObject' (java:8938): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed (java:8938): GLib-GObject-WARNING **: invalid uninstantiatable type '' in cast to 'GObject' (java:8938): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed Are the debug packages available anywhere at http://debug.mirrors.debian.org/? If so, I can install them and dig a bit more. ATB, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 19/02/18 08:39, Samuel Thibault wrote: Hello, Mark Cave-Ayland, on lun. 19 févr. 2018 08:14:13 +, wrote: I've also attached another trace from gdb by doing the following to catch the sources of the above errors emitted on the console: break g_log if log_level == G_LOG_LEVEL_CRITICAL break g_log if log_level == G_LOG_LEVEL_WARNING handle SIGSEGV nostop noprint pass Thanks! The corresponding backtrace look like these warnings are probably fixed by the -16 upload too. No worries - I have a full day of Java development ahead of me, so I will install your -16 package now and see how I go. Thanks a lot for taking the time to look into this, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 17/02/18 17:49, Samuel Thibault wrote: Hello, Mark Cave-Ayland, on ven. 16 févr. 2018 13:46:36 +, wrote: As requested, please find attached a .xz file containing the output of "thread apply all bt full" for the crashing process after installing the debugging symbols using the instructions above. Thanks! Hopefully this will enable this bug to finally be tracked down and fixed. That made me find one issue that is very possibly related indeed. Could you try libatk-wrapper-java-jni_0.33.3-16 which I have just uploaded? Fantastic! I've just submitted some extra traces for the console warnings above which I hope you will find helpful. Poking around a bit with gdb it appears that the g_hash_table entries are coming back as NULL for both the key and entry, so let me know if there's any other info I can provide. Do you still want me to go ahead and install the libatk-wrapper-java-jni_0.33.3-16 package to test your latest fixes, or would you like to review my latest gdb trace first just in case you need any further info? ATB, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
On 17/02/18 21:18, Samuel Thibault wrote: Hello, Mark Cave-Ayland, on ven. 16 févr. 2018 17:18:52 +, wrote: One other thing to note: if I launch netbeans from a terminal I see a constant stream of messages like this appearing whilst I work: (java:23186): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed (java:23186): GLib-GObject-WARNING **: invalid uninstantiatable type '(null)' in cast to 'AtkObject' (java:23186): GLib-GObject-WARNING **: instance of invalid non-instantiatable type '(null)' (java:23186): GLib-GObject-CRITICAL **: g_signal_emit_by_name: assertion 'G_TYPE_CHECK_INSTANCE (instance)' failed (java:23186): GLib-GObject-WARNING **: invalid uninstantiatable type '(null)' in cast to 'GObject' (java:23186): GLib-GObject-WARNING **: invalid uninstantiatable type '(null)' in cast to 'GObject' (java:23186): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed (this is repeated several hundred times) It would be useful to know what actions trigger these (since I've been unable to reproduce them, but I'm not a netbeans user) Hi Samuel, Let me try and explain a bit about my setup: it's an up-to-date Debian stretch amd64 installation with netbeans 8.2 downloaded from netbeans.org. I'm currently writing a spring boot application, and so I have installed the latest version of the gradle and git plugins. My project consists of 3 projects: core, client and web. core and client are gradle JAR subprojects that have shared namespaces, e.g. core could contain org.foo.CoreObject whilst client may contain org.foo.ClientCoreObject. The web application is the root application, and depends upon both the core and client JARs. Once the root web application has been loaded, I can trigger the bug quite easily by using the autocomplete in the IDE for one of my classes, e.g. typing ClassName. (wait for popup list of options, then hit CTRL-space) And if I keep doing this a few times then I can cause the crash. I've also attached another trace from gdb by doing the following to catch the sources of the above errors emitted on the console: break g_log if log_level == G_LOG_LEVEL_CRITICAL break g_log if log_level == G_LOG_LEVEL_WARNING handle SIGSEGV nostop noprint pass ATB, Mark. openjdk-8-8u151-b12-1~deb9u1-netbeans-8.2-glib-warnings.txt.xz Description: application/xz
Bug#837081: netbeans: Crashes due to assertion failure in GLib
One other thing to note: if I launch netbeans from a terminal I see a constant stream of messages like this appearing whilst I work: (java:23186): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed (java:23186): GLib-GObject-WARNING **: invalid uninstantiatable type '(null)' in cast to 'AtkObject' (java:23186): GLib-GObject-WARNING **: instance of invalid non-instantiatable type '(null)' (java:23186): GLib-GObject-CRITICAL **: g_signal_emit_by_name: assertion 'G_TYPE_CHECK_INSTANCE (instance)' failed (java:23186): GLib-GObject-WARNING **: invalid uninstantiatable type '(null)' in cast to 'GObject' (java:23186): GLib-GObject-WARNING **: invalid uninstantiatable type '(null)' in cast to 'GObject' (java:23186): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed (this is repeated several hundred times) ATB, Mark.
Bug#837081: netbeans: Crashes due to assertion failure in GLib
Hi all, For some reason after recently updating from a traditional HD to an SSD, I'm now getting this on a regular basis developing with Netbeans 8.2 on Debian Stretch. As requested, please find attached a .xz file containing the output of "thread apply all bt full" for the crashing process after installing the debugging symbols using the instructions above. Hopefully this will enable this bug to finally be tracked down and fixed. ATB, Mark. openjdk-8-8u151-b12-1~deb9u1-netbeans-8.2-backtrace.txt.xz Description: application/xz
Bug#658886: Removing proll from Debian?
On 04/10/16 22:17, Adrian Bunk wrote: > Hi, > > proll is a JavaStation PROM replacement. > > JavaStations are 32bit, so not supported by the new sparc64 port. > The old 32bit sparc port has already been remved from Debian. > > Guillem mentions in #658886 that QEMU once used to use proll > (but didn't anymore in 2012?). > > Am I right that proll is no longer of use for anyone and should be > removed from unstable, or are there any usecases left? > > cu > Adrian Hi Adrian, QEMU switched to OpenBIOS from proll several years ago, so the QEMU dependency is no longer relevant today. ATB, Mark.
Bug#824665: openbios-sparc: missing QEMU,cgthree.bin file
Note that the QEMU,cgthree.bin binary was first added to QEMU git master in October 2013 in time for the 2.0 release so this bug actually affects all of -backports, -stable, -testing and -unstable packages. ATB, Mark.
Bug#782620: tightvncserver: SIGSEGV in WaitForSomething
I just experienced this today and found that the patch below fixed the problem for me: http://sourceforge.net/p/vnc-tight/patches/90/ Any chance that this could be applied as a fix for wheezy/jessie?
Bug#782620: tightvncserver: SIGSEGV in WaitForSomething
On 24/01/16 19:10, Ola Lundqvist wrote: > tag 782620 + patch > thanks > > Hi Mark > > Thanks for the report. > I can not, as a maintainer, judge that. It has to be decided by the > release team. However I do not think it will be accepted. It is > essentially just security problems that are fixed in stable releases. > > Wheezy is also old-stable and only get critical security fixes, to my > knowledge. That's a real shame as tightvncserver is basically unusable on amd64 (and probably any 64-bit platform) without this patch with constant segfaults during use - fingers crossed that this can be put forward for consideration. > Thanks anyway for the patch. I'll try to get it to unstable+testing at > least. :-) Great - thanks a lot for your work on this! ATB, Mark.
Bug#764450: Roadmap and Change Log do not correctly inherit versions
Package: mantis Version: 1.2.11-1.2 Severity: important When creating a subproject in Mantis, a bug in the version caching code prevents Roadmap and Change Log from finding and displaying inherited versions. This means that if you use the common scenario of adding versions to the top level project and filing bugs in a subproject, the inherited versions are never found and so the both pages remain empty. The bug has been fixed upstream here: https://github.com/mantisbt/mantisbt/commit/5d79b2785f9bd472d1da84d0e26950b3f98639d9.patch. Given how common this scenario is, would it be possible to get this patch included into the standard Debian packages? -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#662103: phonon-backend-vlc: Pausing and resuming in Amarok changes volume
I was pleasantly surprised to find that phonon-backend-vlc 0.6.0-1 was pushed to wheezy during my last set of updates, and I can confirm that this new package fixes the issue for me too. Thank you for all your hard work! -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#662103: phonon-backend-vlc: Pausing and resuming in Amarok changes volume to 100%
Hi all, I can confirm that this bug currently still exists in wheezy which on my up-to-date system is still 0.5.0+14.g382da0d-2. Is there any chance of getting this fix backported into wheezy as it is causing my eardrums to suffer on a regular basis :( ATB, Mark. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#618904: [Pkg-openldap-devel] Bug#618904: Slapd hangs when creating new groups with GOsa
On 20/10/11 03:07, Pawel Wilk wrote: Package: slapd Version: 2.4.23-7.2 Severity: important Slapd hangs when creating groups with GOsa. The only way to restart the slapd is then to kill it with SIGKILL and run the service it again. The mentioned groups are RFC2307bis compliant objects based on the object classes: - groupOfNames, - posixGroup. They require at least one member attribute (2.5.4.31) and if there is no member to add (i.e. because group is a POSIX primary group refered only by gidNumber in user objects) then DN of the very same group object is used. So in case of DN: cn=group1,ou=groups,dc=my,dc=org The group will have the initial member set to: cn=group1,ou=groups,dc=my,dc=org I also have memberof module and overlay enabled (with refint) for member and uniqueMember attributes. Based upon my testing above, I eventually deduced that the hang seems to be caused by the presence of the memberOf overlay. Does temporarily removing it from your configuration fix the problem for you? Unfortunately the answer from upstream is upgrade from 2.4.23, so unless Debian can raise an exception to push a version update into squeeze then it's going to require someone to perform a git bisection and backport the fix :( ATB, Mark. -- Mark Cave-Ayland - Senior Technical Architect PostgreSQL - PostGIS Sirius Corporation plc - control through freedom http://www.siriusit.co.uk t: +44 870 608 0063 Sirius Labs: http://www.siriusit.co.uk/labs -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#618904: [Pkg-openldap-devel] Bug#618904: openldap 2.4.23 slapd server process frequently hangs during everyday use
On 23/03/11 19:45, Quanah Gibson-Mount wrote: I would note that you are the only person using OpenLDAP 2.4.23 since it was released on 6/10/2010 to report this issue. So while I concur this is a serious issue for your use of OpenLDAP, it is also somehow related to your specific OpenLDAP configuration. You never provided your configuration in the upstream discussion that I can find, so it's difficult to know what you've done specifically in your environment that is causing the problem to show up. That's true, although no-one really showed any interest after I could verify that 2.4.24 fixed the issue. If you're still interested, I'll see if I can spend some time at the beginning of next week to come up with a reproducible test case. ATB, Mark. -- Mark Cave-Ayland - Senior Technical Architect PostgreSQL - PostGIS Sirius Corporation plc - control through freedom http://www.siriusit.co.uk t: +44 870 608 0063 Sirius Labs: http://www.siriusit.co.uk/labs -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#618904: openldap 2.4.23 slapd server process frequently hangs during everyday use
Package: slapd
Version: 2.4.23-7
Severity: critical
Tags: squeeze
After upgrading our LDAP server from lenny (2.4.11) to squeeze (2.4.23), we
have found
that the slapd process frequently hangs when adding new objects to the LDAP
tree. The
server freezes and will not accept any new connections until it is forcibly
terminated
with kill -9 and then the slapd process restarted.
When the slapd process has hung, I see the following backtrace in gdb:
(gdb) bt full
#0 0x7fa50aca8be5 in pthread_join (threadid=140346751547136,
thread_return=0x0) at pthread_join.c:89
__ignore = value optimized out
_tid = 10340
_buffer = {__routine = 0x7fa50aca8ab0 cleanup, __arg = 0x7fa506457d28,
__canceltype = 105216464, __prev = 0x0}
oldtype = 0
result = value optimized out
#1 0x0042d72c in slapd_daemon () at
/home/devel/openldap/trunk/servers/slapd/daemon.c:2842
listener_tid = 140346751547136
rc = 0
#2 0x0041ae6a in main (argc=9, argv=0x7fffd2f2e5b0) at
/home/devel/openldap/trunk/servers/slapd/main.c:961
i = 9
no_detach = 0
rc = -12
urls = 0x7df0c0 ldap:/// ldapi:///
username = 0x7df100 root
groupname = 0x7df0e0 ldap
sandbox = 0x0
syslogUser = 160
configfile = 0x7df120 /etc/ldap/slapd.conf
configdir = 0x0
serverName = value optimized out
scp = value optimized out
scp_entry = value optimized out
debug_unknowns = 0x0
syslog_unknowns = 0x0
slapd_pid_file_unlink = 1
slapd_args_file_unlink = 1
firstopt = value optimized out
__PRETTY_FUNCTION__ = main
(gdb)
This suggests that there may be some kind of thread locking issue occurring
which is
causing the main process to hang.
I have raised this with upstream here:
http://www.openldap.org/lists/openldap-technical/201103/msg00175.html
and confirmed that the same issue occurs on a vanilla build of 2.4.23. Further
testing
here shows that issue appears to be resolved in 2.4.24.
This is an extremely critical DOS issue and means that Squeeze's 2.4.23
packages are
unusable in a production environment.
-- System Information:
Debian Release: 6.0
APT prefers stable
APT policy: (500, 'stable')
Architecture: amd64 (x86_64)
Kernel: Linux 2.6.32-5-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Versions of packages slapd depends on:
ii adduser 3.112+nmu2 add and remove users and groups
ii coreutils 8.5-1GNU core utilities
ii debconf [debconf-2.0] 1.5.36.1 Debian configuration management sy
ii libc6 2.11.2-10Embedded GNU C Library: Shared lib
ii libdb4.84.8.30-2 Berkeley v4.8 Database Libraries [
ii libgnutls26 2.8.6-1 the GNU TLS library - runtime libr
ii libldap-2.4-2 2.4.23-7 OpenLDAP libraries
ii libltdl72.2.6b-2 A system independent dlopen wrappe
ii libperl5.10 5.10.1-17shared Perl library
ii libsasl2-2 2.1.23.dfsg1-7 Cyrus SASL - authentication abstra
ii libslp1 1.2.1-7.8OpenSLP libraries
ii libuuid12.17.2-9 Universally Unique ID library
ii libwrap07.6.q-19 Wietse Venema's TCP wrappers libra
ii lsb-base3.2-23.2squeeze1 Linux Standard Base 3.2 init scrip
ii perl [libmime-base64-pe 5.10.1-17Larry Wall's Practical Extraction
ii psmisc 22.11-1 utilities that use the proc file s
ii unixodbc2.2.14p2-1 ODBC tools libraries
Versions of packages slapd recommends:
ii libsasl2-modules 2.1.23.dfsg1-7 Cyrus SASL - pluggable authenticat
Versions of packages slapd suggests:
ii ldap-utils2.4.23-7 OpenLDAP utilities
-- debconf information excluded
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
