[users] yum segmentation fault with mrepo

[Eric Baplue]

Hello,

I have running an mrepo (0.8.7-1.el5.rf) server on centos (5.4). From
here I update my RHEL server with yum. One of the reposytries I mirror
is rpmforge.

Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

I have check several servers and they all throw the message at the same moment.

When I rebuild the mrepo (mrepo -gvvv rhel5Server) I found

8568/10237 - root-tail-1.2-1.el5.rf.i386.rpm
8569/10237 - perl-BSD-Resource-1.2903-1.el5.rf.i386.rpm
8570/10237 - libarxx-0.7.8-1.el5.rf.i386.rpm
8571/10237 - perl-WWW-Search-2.494-1.el5.rf.noarch.rpm
8572/10237 - perl-Parse-RecDescent-1.962.2-1.el5.rf.noarch.rpm

I removed them from the dist, deleted the www and did an update/rebuild from the repository.
I did a "yum clean all".

Whathever I do, I still receive the error on exact the same moment.

Is this a yum problem? a mrepo issue? Where can I look further?

Thx.

Eric Baplue

[Pavel Kankovsky]

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

Thanks for the response but how do I do this, it seems that coredumps
are only for (kernel) deamons and gdb only for C programmes.

Post by Pavel Kankovsky

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

The problem is solved, I have rebuild the whole mrepo www directory and
now yum works fine.

Post by Pavel Kankovsky

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

Thanks for the response but how do I do this, it seems that coredumps
are only for (kernel) deamons and gdb only for C programmes.

Post by Pavel Kankovsky

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

The problem is solved, I have rebuild the whole mrepo www directory and
now yum works fine.

Post by Pavel Kankovsky

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

Thanks for the response but how do I do this, it seems that coredumps
are only for (kernel) deamons and gdb only for C programmes.

Post by Pavel Kankovsky

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

The problem is solved, I have rebuild the whole mrepo www directory and
now yum works fine.

Post by Pavel Kankovsky

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

Thanks for the response but how do I do this, it seems that coredumps
are only for (kernel) deamons and gdb only for C programmes.

Post by Pavel Kankovsky

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

The problem is solved, I have rebuild the whole mrepo www directory and
now yum works fine.

Post by Pavel Kankovsky

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

Thanks for the response but how do I do this, it seems that coredumps
are only for (kernel) deamons and gdb only for C programmes.

Post by Pavel Kankovsky

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

The problem is solved, I have rebuild the whole mrepo www directory and
now yum works fine.

Post by Pavel Kankovsky

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

Hello,

I have running an mrepo (0.8.7-1.el5.rf) server on centos (5.4). From
here I update my RHEL server with yum. One of the reposytries I mirror
is rpmforge.

Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

I have check several servers and they all throw the message at the same moment.

When I rebuild the mrepo (mrepo -gvvv rhel5Server) I found

8568/10237 - root-tail-1.2-1.el5.rf.i386.rpm
8569/10237 - perl-BSD-Resource-1.2903-1.el5.rf.i386.rpm
8570/10237 - libarxx-0.7.8-1.el5.rf.i386.rpm
8571/10237 - perl-WWW-Search-2.494-1.el5.rf.noarch.rpm
8572/10237 - perl-Parse-RecDescent-1.962.2-1.el5.rf.noarch.rpm

I removed them from the dist, deleted the www and did an update/rebuild from the repository.
I did a "yum clean all".

Whathever I do, I still receive the error on exact the same moment.

Is this a yum problem? a mrepo issue? Where can I look further?

Thx.

Eric Baplue

[Pavel Kankovsky]

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

Hello,

I have running an mrepo (0.8.7-1.el5.rf) server on centos (5.4). From
here I update my RHEL server with yum. One of the reposytries I mirror
is rpmforge.

Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

I have check several servers and they all throw the message at the same moment.

When I rebuild the mrepo (mrepo -gvvv rhel5Server) I found

8568/10237 - root-tail-1.2-1.el5.rf.i386.rpm
8569/10237 - perl-BSD-Resource-1.2903-1.el5.rf.i386.rpm
8570/10237 - libarxx-0.7.8-1.el5.rf.i386.rpm
8571/10237 - perl-WWW-Search-2.494-1.el5.rf.noarch.rpm
8572/10237 - perl-Parse-RecDescent-1.962.2-1.el5.rf.noarch.rpm

I removed them from the dist, deleted the www and did an update/rebuild from the repository.
I did a "yum clean all".

Whathever I do, I still receive the error on exact the same moment.

Is this a yum problem? a mrepo issue? Where can I look further?

Thx.

Eric Baplue

[Pavel Kankovsky]

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

Hello,

I have running an mrepo (0.8.7-1.el5.rf) server on centos (5.4). From
here I update my RHEL server with yum. One of the reposytries I mirror
is rpmforge.

Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

I have check several servers and they all throw the message at the same moment.

When I rebuild the mrepo (mrepo -gvvv rhel5Server) I found

8568/10237 - root-tail-1.2-1.el5.rf.i386.rpm
8569/10237 - perl-BSD-Resource-1.2903-1.el5.rf.i386.rpm
8570/10237 - libarxx-0.7.8-1.el5.rf.i386.rpm
8571/10237 - perl-WWW-Search-2.494-1.el5.rf.noarch.rpm
8572/10237 - perl-Parse-RecDescent-1.962.2-1.el5.rf.noarch.rpm

I removed them from the dist, deleted the www and did an update/rebuild from the repository.
I did a "yum clean all".

Whathever I do, I still receive the error on exact the same moment.

Is this a yum problem? a mrepo issue? Where can I look further?

Thx.

Eric Baplue

[Pavel Kankovsky]

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Eric Baplue]

Hello,

I have running an mrepo (0.8.7-1.el5.rf) server on centos (5.4). From
here I update my RHEL server with yum. One of the reposytries I mirror
is rpmforge.

Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

I have check several servers and they all throw the message at the same moment.

When I rebuild the mrepo (mrepo -gvvv rhel5Server) I found

8568/10237 - root-tail-1.2-1.el5.rf.i386.rpm
8569/10237 - perl-BSD-Resource-1.2903-1.el5.rf.i386.rpm
8570/10237 - libarxx-0.7.8-1.el5.rf.i386.rpm
8571/10237 - perl-WWW-Search-2.494-1.el5.rf.noarch.rpm
8572/10237 - perl-Parse-RecDescent-1.962.2-1.el5.rf.noarch.rpm

I removed them from the dist, deleted the www and did an update/rebuild from the repository.
I did a "yum clean all".

Whathever I do, I still receive the error on exact the same moment.

Is this a yum problem? a mrepo issue? Where can I look further?

Thx.

Eric Baplue

[Pavel Kankovsky]

Post by Eric Baplue
Since a few day's I receive from the RHEL server an error.
When I do
"yum update -y --downloadonly"
manualy (this is normaly done from the cron once a week) I get

rhel5Server-servsup.rpmforge | 951 B 00:00
rhel5Server-servsup.rpmforge/primary | 3.6 MB 00:00
rhel5Server-servsup.rpmforge: [############################################################################################## ] 8570/10237Segmentation fault

It happens when the repo cache db is being rebuilt from a downloaded XML?
Enable coredumps or run yum under gdb and get a stack trace.

[Pavel Kankovsky]

Post by Eric Baplue
Thanks for the response but how do I do this, it seems that coredumps
are only for (kernel) deamons and gdb only for C programmes.

You can certainly get a coredump for any process (although it might be
necessary to set ulimit -c or avoid setuid/setgid during execution) or run
it under gdb (after all, Python interpreter *is* a C program).

Post by Eric Baplue
The problem is solved, I have rebuild the whole mrepo www directory and
now yum works fine.

That does not fix the bug that made Yum crash. You did not save the old
broken crash-inducing repo, did you?

[Pavel Kankovsky]

Post by Eric Baplue
Thanks for the response but how do I do this, it seems that coredumps
are only for (kernel) deamons and gdb only for C programmes.

You can certainly get a coredump for any process (although it might be
necessary to set ulimit -c or avoid setuid/setgid during execution) or run
it under gdb (after all, Python interpreter *is* a C program).

Post by Eric Baplue
The problem is solved, I have rebuild the whole mrepo www directory and
now yum works fine.

That does not fix the bug that made Yum crash. You did not save the old
broken crash-inducing repo, did you?

[Pavel Kankovsky]

Post by Eric Baplue
Thanks for the response but how do I do this, it seems that coredumps
are only for (kernel) deamons and gdb only for C programmes.

You can certainly get a coredump for any process (although it might be
necessary to set ulimit -c or avoid setuid/setgid during execution) or run
it under gdb (after all, Python interpreter *is* a C program).

Post by Eric Baplue
The problem is solved, I have rebuild the whole mrepo www directory and
now yum works fine.

That does not fix the bug that made Yum crash. You did not save the old
broken crash-inducing repo, did you?

[Pavel Kankovsky]

Post by Eric Baplue
Thanks for the response but how do I do this, it seems that coredumps
are only for (kernel) deamons and gdb only for C programmes.

You can certainly get a coredump for any process (although it might be
necessary to set ulimit -c or avoid setuid/setgid during execution) or run
it under gdb (after all, Python interpreter *is* a C program).

Post by Eric Baplue
The problem is solved, I have rebuild the whole mrepo www directory and
now yum works fine.

That does not fix the bug that made Yum crash. You did not save the old
broken crash-inducing repo, did you?

[Pavel Kankovsky]

Post by Eric Baplue
Thanks for the response but how do I do this, it seems that coredumps
are only for (kernel) deamons and gdb only for C programmes.

You can certainly get a coredump for any process (although it might be
necessary to set ulimit -c or avoid setuid/setgid during execution) or run
it under gdb (after all, Python interpreter *is* a C program).

Post by Eric Baplue
The problem is solved, I have rebuild the whole mrepo www directory and
now yum works fine.

That does not fix the bug that made Yum crash. You did not save the old
broken crash-inducing repo, did you?