PDA

View Full Version : Yum Segmentation Fault



narasim_7
19th November 2006, 01:55 PM
Hi..I reinstalled FC5 on my computer yesterday . Today i found that pirut and yumex hung in the middle. Then i ran yum cli only to find out this
yum install yakuake
Loading "installonlyn" plugin
Setting up Install Process
Setting up repositories
dries [1/6]
core [2/6]
updates [3/6]
freshrpms [4/6]
macromedia [5/6]
extras [6/6]
Reading repository metadata in from local files
Segmentation fault

Then I killed yum to run it again , only this time i dont get any segfaults , it never proceeds not even the loading plugin message. I tried this several times after this. Yum would work once after which it will hang.

Any help is appreciated

P4 2.4 Ghz ,845 GVSR ,256 Mb DDR

nlkrio
19th November 2006, 02:23 PM
try
yum clean all
and then
yum install yakuake

narasim_7
19th November 2006, 02:51 PM
I works...Thanks..But still i am not sure..may need a day or two to confirm whether things are ok ..

narasim_7
20th November 2006, 08:47 PM
Problem seems to resolved..I tried with ubuntu 6.06 on the same partition and it stuck while running its package manager, that is , all the packages are downloaded but when it was time to install ,the entire system became unresponsive and finally it rebooted saying "out of memory" . So Installed Fedora core 5 on to another bigger partition and now its ok . :)

narasim_7
21st November 2006, 01:37 PM
Suddenly this happend just a few minutes before posting this.

yum install scribes
Loading "installonlyn" plugin
Setting up Install Process
Setting up repositories
core [1/5]
core 100% |=========================| 1.1 kB 00:00
macromedia [2/5]
macromedia 100% |=========================| 951 B 00:00
updates [3/5]
updates 100% |=========================| 1.2 kB 00:00
freshrpms [4/5]
freshrpms 100% |=========================| 951 B 00:00
extras [5/5]
extras 100% |=========================| 1.1 kB 00:00
Reading repository metadata in from local files
Parsing package install arguments
Resolving Dependencies
--> Populating transaction set with selected packages. Please wait.
---> Downloading header for scribes to pack into transaction set.
scribes-0.2.5-4.fc5.noarc 100% |=========================| 28 kB 00:01
---> Package scribes.noarch 0:0.2.5-4.fc5 set to be updated
--> Running transaction check
Segmentation fault

I dont think that yum clean all will solve the problem. It hangs anyway. After this seg fault occurs i am not able to run yum again . I can do so only after a restart ...And one more thing I have used the alternate configuration of mirros as given in this forum .
Is anyone getting such a problem.?
Kindly help.

P4 2.4 ghz ,Intel 845 GVSR

evans
22nd November 2006, 05:10 PM
I dont think that yum clean all will solve the problem. It hangs anyway. After this seg fault occurs i am not able to run yum again . I can do so only after a restart ...And one more thing I have used the alternate configuration of mirros as given in this forum .
Is anyone getting such a problem.?
Kindly help.

P4 2.4 ghz ,Intel 845 GVSR


I just got the same problem, using yum-3.0.1-2.fc6 on a p4 with kernel 2.6.18-1.2849.fc6 i686

Now yum hangs at "loading install pugins".

YTou say a reboot fixes it? Hmm, I never like that soluion. I looked for a pid.lock file, but nothing. Anyone?

narasim_7
22nd November 2006, 06:31 PM
I put the debug level to 3 and watched for the messages after the seg fault..It hangs at RPMDB ..and i killed yum and removed the /var/run/yum.pid and ran rpm command which doesnt work either(it too hangs). Sometimes i even get a rpmdb panic.

Reboot does not fix it..I am able to run yum twice or thrice after which the inevitable happens .. So if i need to use yum i have to restart.

And by the way , You have the problem only on FC6 or in FC5 also?

narasim_7
22nd November 2006, 06:34 PM
I think i have to elaborate on the first line of my above post.. I made the debug level in yum.conf to 3. Then after yum showed a seg fault and quit, i removed /var/run/yum.pid and ran yum again and watched for the messages . It hangs at RPMDB ..and then i killed yum and removed the /var/run/yum.pid and ran rpm command which doesnt work either(it too hangs). Sometimes i even get a rpmdb panic.

evans
22nd November 2006, 06:57 PM
And by the way , You have the problem only on FC6 or in FC5 also?

Sorry, I should have made it clear that I am talking about Fedora Core 6 running on a 2.4GHz P4 Dell Lattitude laptop. I am running the command

rpmdb --rebuilddb
but it taking forever (45 minutes so far) so I think it two is hanging. I will wait in case it will finish, and I do not want to hose my database!

And, no, I do not have this problem on my Fedora Core 5 machines.

beartech
28th November 2006, 05:26 AM
I'm having exactly the same problem
After trying to install yumex via yum I got a segmentation fault


[root@freja ~]# yum -y install yumex
Loading "installonlyn" plugin
Setting up Install Process
Setting up repositories
panorama [1/4]
panorama 100% |=========================| 951 B 00:00
updates [2/4]
updates 100% |=========================| 1.2 kB 00:00
core [3/4]
core 100% |=========================| 1.1 kB 00:00
extras [4/4]
extras 100% |=========================| 1.1 kB 00:00
Reading repository metadata in from local files
primary.xml.gz 100% |=========================| 405 kB 00:07
updates : ################################################## 1118/1118
Added 10 new packages, deleted 10 old in 36.69 seconds
primary.xml.gz 100% |=========================| 1.7 MB 00:30
extras : ################################################## 5740/5740
Added 26 new packages, deleted 23 old in 98.28 seconds
Segmentation fault

I tried removing the pid file at /var/run/yum.pid but all that is happening now is that yum just hangs when I try to do anything with it . The
-e 10 -d 10 options do not seem to be producing any output and the changes made by installing yumex were not recorded in /var/log/yum

Does anyone have any ideas how I can dig myself out of this mess ?

--Simon

brunson
28th November 2006, 05:38 AM
run "strace yum update" and paste us the last 20 or 30 lines of output, please.

narasim_7
2nd December 2006, 06:13 PM
i removed the __db.001 .002 .003 files from /var/lib/rpm and yum starts working again after it hangs..Is this some kind of lock file , can some explain this . Thanks

sr20dem0n
12th December 2006, 06:26 AM
run "strace yum update" and paste us the last 20 or 30 lines of output, please.

open("/var/lib/rpm/__db.002", O_RDWR|O_LARGEFILE) = 9
fcntl64(9, F_SETFD, FD_CLOEXEC) = 0
mmap2(NULL, 1318912, PROT_READ|PROT_WRITE, MAP_SHARED, 9, 0) = 0xb7a8a000
close(9) = 0
stat64("/var/lib/rpm/__db.003", {st_mode=S_IFREG|0644, st_size=450560, ...}) = 0
open("/var/lib/rpm/__db.003", O_RDWR|O_LARGEFILE) = 9
fcntl64(9, F_SETFD, FD_CLOEXEC) = 0
mmap2(NULL, 450560, PROT_READ|PROT_WRITE, MAP_SHARED, 9, 0) = 0xb7a1c000
close(9) = 0
stat64("/var/lib/rpm/Packages", {st_mode=S_IFREG|0644, st_size=47435776, ...}) = 0
open("/var/lib/rpm/Packages", O_RDONLY|O_LARGEFILE) = 9
fcntl64(9, F_SETFD, FD_CLOEXEC) = 0
read(9, "\0\0\0\0\1\0\0\0\0\0\0\0a\25\6\0\10\0\0\0\0\20\0\0 \0\10"..., 512) = 512
close(9) = 0
open("/var/lib/rpm/Packages", O_RDONLY|O_LARGEFILE) = 9
fcntl64(9, F_SETFD, FD_CLOEXEC) = 0
fstat64(9, {st_mode=S_IFREG|0644, st_size=47435776, ...}) = 0
rt_sigprocmask(SIG_BLOCK, ~[RTMIN RT_1], [], 8) = 0
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
futex(0xb7bcac50, FUTEX_WAIT, 2, NULL

and that's where it hangs, at least for me. I'm having the same problem as the others, only without the seg fault, it just hangs forever. Also, if I do a "rpm -q kernel" it prints out the original kernel (2.6.18-1.2798) but then hangs before printing the current kernel (2.6.18-1.2849) - this is FC6

brunson
12th December 2006, 06:43 AM
Same with me, it's waiting for a lock or a semaphore in a memory mapped file. I haven't had the time to download the yum source and try to figure out what's going on. My hangs usually follow a seg fault, so I believe it's the previous failed yum's lock on some of the RPM db files hanging around.

What's odd is that I haven't seen any mention of it on the mailing list. I haven't been bothered enough to bring it up yet, I want more info. But having you corroborate it is great ammunition, thanks for posting.

Removing the rpm database files and rebuilding is only bandaiding the problem, something is broken and needs fixing upstream.

netjrc
9th January 2007, 01:52 AM
I have the exact same problem, the exact same "strace yum update" output (below), and I can solve it temporarily by deleting the /var/lib/rpm/__db.* files.



open("/var/lib/rpm/DB_CONFIG", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory)
stat64("/var/lib/rpm/__db.001", {st_mode=S_IFREG|0644, st_size=24576, ...}) = 0
open("/var/lib/rpm/__db.001", O_RDWR|O_LARGEFILE) = 10
fcntl64(10, F_SETFD, FD_CLOEXEC) = 0
fstat64(10, {st_mode=S_IFREG|0644, st_size=24576, ...}) = 0
close(10) = 0
open("/var/lib/rpm/__db.001", O_RDWR|O_LARGEFILE) = 10
fcntl64(10, F_SETFD, FD_CLOEXEC) = 0
mmap2(NULL, 24576, PROT_READ|PROT_WRITE, MAP_SHARED, 10, 0) = 0xb7b2c000
close(10) = 0
stat64("/var/lib/rpm/__db.002", {st_mode=S_IFREG|0644, st_size=1318912, ...}) = 0
open("/var/lib/rpm/__db.002", O_RDWR|O_LARGEFILE) = 10
fcntl64(10, F_SETFD, FD_CLOEXEC) = 0
mmap2(NULL, 1318912, PROT_READ|PROT_WRITE, MAP_SHARED, 10, 0) = 0xb79ea000
close(10) = 0
stat64("/var/lib/rpm/__db.003", {st_mode=S_IFREG|0644, st_size=450560, ...}) = 0
open("/var/lib/rpm/__db.003", O_RDWR|O_LARGEFILE) = 10
fcntl64(10, F_SETFD, FD_CLOEXEC) = 0
mmap2(NULL, 450560, PROT_READ|PROT_WRITE, MAP_SHARED, 10, 0) = 0xb797c000
close(10) = 0
stat64("/var/lib/rpm/Packages", {st_mode=S_IFREG|0644, st_size=35397632, ...}) = 0
open("/var/lib/rpm/Packages", O_RDONLY|O_LARGEFILE) = 10
fcntl64(10, F_SETFD, FD_CLOEXEC) = 0
read(10, "\0\0\0\0\1\0\0\0\0\0\0\0a\25\6\0\10\0\0\0\0\20\0\0 \0\10"..., 512) = 512
close(10) = 0
open("/var/lib/rpm/Packages", O_RDONLY|O_LARGEFILE) = 10
fcntl64(10, F_SETFD, FD_CLOEXEC) = 0
fstat64(10, {st_mode=S_IFREG|0644, st_size=35397632, ...}) = 0
rt_sigprocmask(SIG_BLOCK, ~[RTMIN RT_1], [], 8) = 0
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
pread64(10, "\0\0\0\0\1\0\0\0\334\0\0\0\0\0\0\0\335\0\0\0\1\0\3 46\17"..., 4096, 901120) = 4096
futex(0xb7b29210, FUTEX_WAIT, 2, NULL

ilbh
9th January 2007, 10:52 AM
if you remove yum and install it again, with a rpm package (yum.rpm)
will it work ?
have you tried ?

leigh123linux
9th January 2007, 10:58 AM
before you reinstall yum try deleting rpmdb the rebuild the complete db from new!!



su -
rm /var/lib/rpm/__db*
rpm --rebuilddb

narasim_7
9th January 2007, 01:02 PM
Now i have fedora core 5 installed and i havent updated my yum yet.. Yum or pirut dont give any such problems now as mentioned by me or others earlier when using FC6.
I would be testing yum with its latest package in FC5 itself

brunson
9th January 2007, 05:57 PM
before you reinstall yum try deleting rpmdb the rebuild the complete db from new!!
That's just treating the symptom, not the problem, you haven't released the lock, you've removed the reference to the locked file and recreated it. You can simply reboot to get rid of the lock, but I consider that just as unacceptable.

I haven't seen the problem in several weeks on my systems, I wonder if yum got fixed upstream and updated.

psyklops
15th January 2007, 11:40 AM
im also seeing this problem on FC5... anyone has an update?

hellsgate
15th January 2007, 11:48 AM
Hy.I'm using FC6. I made yesterday an uodate of the kernel and when ran yum update it won't work.
yum extender doesn't work as well.It says that I have probl;ems with repositories.What is the problem?

narasim_7
21st January 2007, 02:22 PM
The recent yum update (yum-3.0.3-1.fc6) is working without any problems for me. Pirut yumex work fine.

P4 2.4 GHz Intel 845GVSR

FunkyRes
22nd January 2007, 09:50 PM
I'm seeing the same thing on a virgin install of FC6 on an IBM Thinkpad (T20).

The problem happened after importing the gpg key for the first update - and then the segfault.

I was able to yum update yum - but that didn't help.

Since "yum update yum" had worked - I tried something from extras -

yum install gnomesword

seg fault (this is after new yum installed). reboot and same thing.

removed the __db files and rebhuildb - and installing gnomesword worked. Package and dependencies installed.

Will try the yum update again.

starvikas009
20th January 2008, 06:21 AM
Hey I am gettinmg the same problem..........
yum is hanging or going in some deadlock
some times it gives segmentation fault

Yar i am trying to fix it for 2-3 days .. but I cud not... then I visited this thread and thought some more people faced the same problem



[root@localhost platoon2]# yum clean all


yum install Loading "installonlyn" plugin
Cleaning up Everything
yakuake[root@localhost platoon2]#
[root@localhost platoon2]#
[root@localhost platoon2]# yum install yakuake
Loading "installonlyn" plugin
Setting up Install Process
Setting up repositories
core 100% |=========================| 1.1 kB 00:00
updates 100% |=========================| 1.2 kB 00:00
extras 100% |=========================| 1.1 kB 00:00
Reading repository metadata in from local files
primary.xml.gz 100% |=========================| 824 kB 00:12
################################################## 2242/2242
primary.xml.gz 100% |=========================| 475 kB 00:09
################################################## 1285/1285
primary.xml.gz 100% |=========================| 1.7 MB 00:21
################################################## 5211/5211
Parsing package install arguments



su -
rm /var/lib/rpm/__db*
rpm --rebuilddb

it did not help
then
I removed yum

then donload yum.rpm
and install
then also it is giving errors