* [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them
@ 2015-03-09 15:40 Ian Campbell
2015-03-09 21:54 ` Don Slutz
2015-03-24 12:40 ` Ian Jackson
0 siblings, 2 replies; 10+ messages in thread
From: Ian Campbell @ 2015-03-09 15:40 UTC (permalink / raw)
To: ian.jackson; +Cc: Ian Campbell, xen-devel
Refactor the $kvp_replace helper in ts-xen-install into a generic
helper (which requires using ::EO and ::EI for namespacing) for use
with target_editfile and use it to edit /etc/sysctl.conf to set
kernel.core_pattern on boot.
Tested in standalone mode by installing and running a C program
containing "*(int *)0 = 1;" which, after running "ulimit -c unlimited"
produces the expected core file. ts-logs-capture when run in
standalone mode then picks them up.
I've not yet figured out how to make the desired rlimit take affect
for all processes (including e.g. daemons spawned on boot). Likely
this will involve some combination of pam_limits.so PAM module and
adding explicit ulimit calls to the initscripts which we care about
(primarily xencommons and libvirt initscripts).
I did debate making the presence of cores in /var/core fail the test
(somehow), but I decided that would be annoying for standalone mode or
shared host scenarios where the core files might be stale or related
to another job.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
Osstest/TestSupport.pm | 22 ++++++++++++++++++++++
ts-host-install | 9 +++++++++
ts-logs-capture | 2 ++
ts-xen-install | 19 ++-----------------
4 files changed, 35 insertions(+), 17 deletions(-)
diff --git a/Osstest/TestSupport.pm b/Osstest/TestSupport.pm
index 8754e22..ece2282 100644
--- a/Osstest/TestSupport.pm
+++ b/Osstest/TestSupport.pm
@@ -57,6 +57,7 @@ BEGIN {
target_put_guest_image target_editfile
target_editfile_cancel
target_editfile_root target_file_exists
+ target_editfile_kvp_replace
target_run_apt
target_install_packages target_install_packages_norec
target_jobdir target_extract_jobdistpath_subdir
@@ -542,6 +543,27 @@ sub teditfileex {
if $install;
}
+# Replace a Key=Value style line in a config file.
+#
+# To be used as 3rd argument to target_editfile(_root) as:
+# target_editfile_root($ho, "/path/to/a/file",
+# sub { target_editfile_kvp_replace($key, $value) });
+sub target_editfile_kvp_replace ($$)
+{
+ my ($key,$value) = @_;
+ my $prnow;
+ $prnow= sub {
+ print ::EO "$key=$value\n" or die $!;
+ $prnow= sub { };
+ };
+ while (<::EI>) {
+ print ::EO or die $! unless m/^$key\b/;
+ $prnow->() if m/^#$key/;
+ }
+ print ::EO "\n" or die $!;
+ $prnow->();
+};
+
sub target_editfile_root ($$$;$$) { teditfileex('root',@_); }
sub target_editfile ($$$;$$) { teditfileex('osstest',@_); }
# my $code= pop @_;
diff --git a/ts-host-install b/ts-host-install
index 9656079..b60abae 100755
--- a/ts-host-install
+++ b/ts-host-install
@@ -139,6 +139,15 @@ END
});
}
+ target_cmd_root($ho, 'mkdir -p /var/core');
+
+ target_editfile_root($ho, '/etc/sysctl.conf',
+ sub { target_editfile_kvp_replace(
+ "kernel.core_pattern",
+ # %p==pid,%e==executable name,%t==timestamp
+ "/var/core/%t.%p.%e.core") });
+ target_cmd_root($ho, "sysctl --load /etc/sysctl.conf");
+
target_cmd_root($ho, "update-rc.d osstest-confirm-booted start 99 2 .");
logm('OK: install completed');
diff --git a/ts-logs-capture b/ts-logs-capture
index 453b03d..45b0a38 100755
--- a/ts-logs-capture
+++ b/ts-logs-capture
@@ -136,6 +136,8 @@ sub fetch_logs_host_guests () {
/home/osstest/osstest-confirm-booted.log
+ /var/core/*.core
+
)];
if (!try_fetch_logs($ho, $logs)) {
logm("log fetching failed, trying hard host reboot...");
diff --git a/ts-xen-install b/ts-xen-install
index 5282f0a..da64a90 100755
--- a/ts-xen-install
+++ b/ts-xen-install
@@ -114,26 +114,11 @@ sub adjustconfig () {
}
die unless defined $trace_config_file;
- my $kvp_replace = sub($$) {
- my ($key,$value) = @_;
- my $prnow;
- $prnow= sub {
- print EO "$key=$value\n" or die $!;
- $prnow= sub { };
- };
- while (<EI>) {
- print EO or die $! unless m/^$key\b/;
- $prnow->() if m/^#$key/;
- }
- print EO "\n" or die $!;
- $prnow->();
- };
-
target_editfile_root($ho, $trace_config_file,
- sub { $kvp_replace->("XENCONSOLED_TRACE", "guest") });
+ sub { target_editfile_kvp_replace("XENCONSOLED_TRACE", "guest") });
target_editfile_root($ho, '/etc/libvirt/libvirtd.conf',
- sub { $kvp_replace->("log_level", "1") })
+ sub { target_editfile_kvp_replace("log_level", "1") })
if toolstack($ho)->{Name} eq "libvirt";
target_cmd_root($ho, 'mkdir -p /var/log/xen/console');
--
2.1.4
^ permalink raw reply related [flat|nested] 10+ messages in thread* Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them
2015-03-09 15:40 [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them Ian Campbell
@ 2015-03-09 21:54 ` Don Slutz
2015-03-10 9:38 ` Ian Campbell
2015-03-24 12:40 ` Ian Jackson
1 sibling, 1 reply; 10+ messages in thread
From: Don Slutz @ 2015-03-09 21:54 UTC (permalink / raw)
To: Ian Campbell, ian.jackson; +Cc: xen-devel
On 03/09/15 11:40, Ian Campbell wrote:
> Refactor the $kvp_replace helper in ts-xen-install into a generic
> helper (which requires using ::EO and ::EI for namespacing) for use
> with target_editfile and use it to edit /etc/sysctl.conf to set
> kernel.core_pattern on boot.
>
> Tested in standalone mode by installing and running a C program
> containing "*(int *)0 = 1;" which, after running "ulimit -c unlimited"
> produces the expected core file. ts-logs-capture when run in
> standalone mode then picks them up.
>
> I've not yet figured out how to make the desired rlimit take affect
> for all processes (including e.g. daemons spawned on boot). Likely
> this will involve some combination of pam_limits.so PAM module and
> adding explicit ulimit calls to the initscripts which we care about
> (primarily xencommons and libvirt initscripts).
I am not sure about debian, but for fedora the places are:
/etc/security/limits.conf:
* soft core unlimited
/etc/profile:
ulimit -S -c unlimited > /dev/null 2>&1
/etc/sysctl.conf
fs.suid_dumpable = 1
/etc/sysconfig/init:
DAEMON_COREFILE_LIMIT='unlimited'
Note: The last depends on:
/etc/init.d/functions:
ulimit -S -c ${DAEMON_COREFILE_LIMIT:-0} >/dev/null 2>&1
Hope this helps.
-Don Slutz
>
> I did debate making the presence of cores in /var/core fail the test
> (somehow), but I decided that would be annoying for standalone mode or
> shared host scenarios where the core files might be stale or related
> to another job.
>
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
> ---
> Osstest/TestSupport.pm | 22 ++++++++++++++++++++++
> ts-host-install | 9 +++++++++
> ts-logs-capture | 2 ++
> ts-xen-install | 19 ++-----------------
> 4 files changed, 35 insertions(+), 17 deletions(-)
>
> diff --git a/Osstest/TestSupport.pm b/Osstest/TestSupport.pm
> index 8754e22..ece2282 100644
> --- a/Osstest/TestSupport.pm
> +++ b/Osstest/TestSupport.pm
> @@ -57,6 +57,7 @@ BEGIN {
> target_put_guest_image target_editfile
> target_editfile_cancel
> target_editfile_root target_file_exists
> + target_editfile_kvp_replace
> target_run_apt
> target_install_packages target_install_packages_norec
> target_jobdir target_extract_jobdistpath_subdir
> @@ -542,6 +543,27 @@ sub teditfileex {
> if $install;
> }
>
> +# Replace a Key=Value style line in a config file.
> +#
> +# To be used as 3rd argument to target_editfile(_root) as:
> +# target_editfile_root($ho, "/path/to/a/file",
> +# sub { target_editfile_kvp_replace($key, $value) });
> +sub target_editfile_kvp_replace ($$)
> +{
> + my ($key,$value) = @_;
> + my $prnow;
> + $prnow= sub {
> + print ::EO "$key=$value\n" or die $!;
> + $prnow= sub { };
> + };
> + while (<::EI>) {
> + print ::EO or die $! unless m/^$key\b/;
> + $prnow->() if m/^#$key/;
> + }
> + print ::EO "\n" or die $!;
> + $prnow->();
> +};
> +
> sub target_editfile_root ($$$;$$) { teditfileex('root',@_); }
> sub target_editfile ($$$;$$) { teditfileex('osstest',@_); }
> # my $code= pop @_;
> diff --git a/ts-host-install b/ts-host-install
> index 9656079..b60abae 100755
> --- a/ts-host-install
> +++ b/ts-host-install
> @@ -139,6 +139,15 @@ END
> });
> }
>
> + target_cmd_root($ho, 'mkdir -p /var/core');
> +
> + target_editfile_root($ho, '/etc/sysctl.conf',
> + sub { target_editfile_kvp_replace(
> + "kernel.core_pattern",
> + # %p==pid,%e==executable name,%t==timestamp
> + "/var/core/%t.%p.%e.core") });
> + target_cmd_root($ho, "sysctl --load /etc/sysctl.conf");
> +
> target_cmd_root($ho, "update-rc.d osstest-confirm-booted start 99 2 .");
>
> logm('OK: install completed');
> diff --git a/ts-logs-capture b/ts-logs-capture
> index 453b03d..45b0a38 100755
> --- a/ts-logs-capture
> +++ b/ts-logs-capture
> @@ -136,6 +136,8 @@ sub fetch_logs_host_guests () {
>
> /home/osstest/osstest-confirm-booted.log
>
> + /var/core/*.core
> +
> )];
> if (!try_fetch_logs($ho, $logs)) {
> logm("log fetching failed, trying hard host reboot...");
> diff --git a/ts-xen-install b/ts-xen-install
> index 5282f0a..da64a90 100755
> --- a/ts-xen-install
> +++ b/ts-xen-install
> @@ -114,26 +114,11 @@ sub adjustconfig () {
> }
> die unless defined $trace_config_file;
>
> - my $kvp_replace = sub($$) {
> - my ($key,$value) = @_;
> - my $prnow;
> - $prnow= sub {
> - print EO "$key=$value\n" or die $!;
> - $prnow= sub { };
> - };
> - while (<EI>) {
> - print EO or die $! unless m/^$key\b/;
> - $prnow->() if m/^#$key/;
> - }
> - print EO "\n" or die $!;
> - $prnow->();
> - };
> -
> target_editfile_root($ho, $trace_config_file,
> - sub { $kvp_replace->("XENCONSOLED_TRACE", "guest") });
> + sub { target_editfile_kvp_replace("XENCONSOLED_TRACE", "guest") });
>
> target_editfile_root($ho, '/etc/libvirt/libvirtd.conf',
> - sub { $kvp_replace->("log_level", "1") })
> + sub { target_editfile_kvp_replace("log_level", "1") })
> if toolstack($ho)->{Name} eq "libvirt";
>
> target_cmd_root($ho, 'mkdir -p /var/log/xen/console');
>
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them
2015-03-09 21:54 ` Don Slutz
@ 2015-03-10 9:38 ` Ian Campbell
0 siblings, 0 replies; 10+ messages in thread
From: Ian Campbell @ 2015-03-10 9:38 UTC (permalink / raw)
To: Don Slutz; +Cc: ian.jackson, xen-devel
On Mon, 2015-03-09 at 17:54 -0400, Don Slutz wrote:
>
> On 03/09/15 11:40, Ian Campbell wrote:
> > Refactor the $kvp_replace helper in ts-xen-install into a generic
> > helper (which requires using ::EO and ::EI for namespacing) for use
> > with target_editfile and use it to edit /etc/sysctl.conf to set
> > kernel.core_pattern on boot.
> >
> > Tested in standalone mode by installing and running a C program
> > containing "*(int *)0 = 1;" which, after running "ulimit -c unlimited"
> > produces the expected core file. ts-logs-capture when run in
> > standalone mode then picks them up.
> >
> > I've not yet figured out how to make the desired rlimit take affect
> > for all processes (including e.g. daemons spawned on boot). Likely
> > this will involve some combination of pam_limits.so PAM module and
> > adding explicit ulimit calls to the initscripts which we care about
> > (primarily xencommons and libvirt initscripts).
>
> I am not sure about debian, but for fedora the places are:
Thanks, these look broadly similar to Debian.
> /etc/security/limits.conf:
> * soft core unlimited
FYI "*" explicitly excludes root, at least on Debian, so a separate line
would be needed for root logins.
> /etc/profile:
> ulimit -S -c unlimited > /dev/null 2>&1
>
> /etc/sysctl.conf
> fs.suid_dumpable = 1
Not 100% sure if we would need this in the context of osstest.
> /etc/sysconfig/init:
> DAEMON_COREFILE_LIMIT='unlimited'
>
> Note: The last depends on:
> /etc/init.d/functions:
> ulimit -S -c ${DAEMON_COREFILE_LIMIT:-0} >/dev/null 2>&1
I think this one is a Red Hat ism.
> Hope this helps.
It did, thanks.
> -Don Slutz
>
> >
> > I did debate making the presence of cores in /var/core fail the test
> > (somehow), but I decided that would be annoying for standalone mode or
> > shared host scenarios where the core files might be stale or related
> > to another job.
> >
> > Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
> > ---
> > Osstest/TestSupport.pm | 22 ++++++++++++++++++++++
> > ts-host-install | 9 +++++++++
> > ts-logs-capture | 2 ++
> > ts-xen-install | 19 ++-----------------
> > 4 files changed, 35 insertions(+), 17 deletions(-)
> >
> > diff --git a/Osstest/TestSupport.pm b/Osstest/TestSupport.pm
> > index 8754e22..ece2282 100644
> > --- a/Osstest/TestSupport.pm
> > +++ b/Osstest/TestSupport.pm
> > @@ -57,6 +57,7 @@ BEGIN {
> > target_put_guest_image target_editfile
> > target_editfile_cancel
> > target_editfile_root target_file_exists
> > + target_editfile_kvp_replace
> > target_run_apt
> > target_install_packages target_install_packages_norec
> > target_jobdir target_extract_jobdistpath_subdir
> > @@ -542,6 +543,27 @@ sub teditfileex {
> > if $install;
> > }
> >
> > +# Replace a Key=Value style line in a config file.
> > +#
> > +# To be used as 3rd argument to target_editfile(_root) as:
> > +# target_editfile_root($ho, "/path/to/a/file",
> > +# sub { target_editfile_kvp_replace($key, $value) });
> > +sub target_editfile_kvp_replace ($$)
> > +{
> > + my ($key,$value) = @_;
> > + my $prnow;
> > + $prnow= sub {
> > + print ::EO "$key=$value\n" or die $!;
> > + $prnow= sub { };
> > + };
> > + while (<::EI>) {
> > + print ::EO or die $! unless m/^$key\b/;
> > + $prnow->() if m/^#$key/;
> > + }
> > + print ::EO "\n" or die $!;
> > + $prnow->();
> > +};
> > +
> > sub target_editfile_root ($$$;$$) { teditfileex('root',@_); }
> > sub target_editfile ($$$;$$) { teditfileex('osstest',@_); }
> > # my $code= pop @_;
> > diff --git a/ts-host-install b/ts-host-install
> > index 9656079..b60abae 100755
> > --- a/ts-host-install
> > +++ b/ts-host-install
> > @@ -139,6 +139,15 @@ END
> > });
> > }
> >
> > + target_cmd_root($ho, 'mkdir -p /var/core');
> > +
> > + target_editfile_root($ho, '/etc/sysctl.conf',
> > + sub { target_editfile_kvp_replace(
> > + "kernel.core_pattern",
> > + # %p==pid,%e==executable name,%t==timestamp
> > + "/var/core/%t.%p.%e.core") });
> > + target_cmd_root($ho, "sysctl --load /etc/sysctl.conf");
> > +
> > target_cmd_root($ho, "update-rc.d osstest-confirm-booted start 99 2 .");
> >
> > logm('OK: install completed');
> > diff --git a/ts-logs-capture b/ts-logs-capture
> > index 453b03d..45b0a38 100755
> > --- a/ts-logs-capture
> > +++ b/ts-logs-capture
> > @@ -136,6 +136,8 @@ sub fetch_logs_host_guests () {
> >
> > /home/osstest/osstest-confirm-booted.log
> >
> > + /var/core/*.core
> > +
> > )];
> > if (!try_fetch_logs($ho, $logs)) {
> > logm("log fetching failed, trying hard host reboot...");
> > diff --git a/ts-xen-install b/ts-xen-install
> > index 5282f0a..da64a90 100755
> > --- a/ts-xen-install
> > +++ b/ts-xen-install
> > @@ -114,26 +114,11 @@ sub adjustconfig () {
> > }
> > die unless defined $trace_config_file;
> >
> > - my $kvp_replace = sub($$) {
> > - my ($key,$value) = @_;
> > - my $prnow;
> > - $prnow= sub {
> > - print EO "$key=$value\n" or die $!;
> > - $prnow= sub { };
> > - };
> > - while (<EI>) {
> > - print EO or die $! unless m/^$key\b/;
> > - $prnow->() if m/^#$key/;
> > - }
> > - print EO "\n" or die $!;
> > - $prnow->();
> > - };
> > -
> > target_editfile_root($ho, $trace_config_file,
> > - sub { $kvp_replace->("XENCONSOLED_TRACE", "guest") });
> > + sub { target_editfile_kvp_replace("XENCONSOLED_TRACE", "guest") });
> >
> > target_editfile_root($ho, '/etc/libvirt/libvirtd.conf',
> > - sub { $kvp_replace->("log_level", "1") })
> > + sub { target_editfile_kvp_replace("log_level", "1") })
> > if toolstack($ho)->{Name} eq "libvirt";
> >
> > target_cmd_root($ho, 'mkdir -p /var/log/xen/console');
> >
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them
2015-03-09 15:40 [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them Ian Campbell
2015-03-09 21:54 ` Don Slutz
@ 2015-03-24 12:40 ` Ian Jackson
2015-03-24 12:46 ` Ian Campbell
1 sibling, 1 reply; 10+ messages in thread
From: Ian Jackson @ 2015-03-24 12:40 UTC (permalink / raw)
To: Ian Campbell; +Cc: xen-devel
Ian Campbell writes ("[PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them"):
> Refactor the $kvp_replace helper in ts-xen-install into a generic
> helper (which requires using ::EO and ::EI for namespacing) for use
> with target_editfile [etc.]
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
> I did debate making the presence of cores in /var/core fail the test
> (somehow), but I decided that would be annoying for standalone mode or
> shared host scenarios where the core files might be stale or related
> to another job.
I think you should add /var/core to the leak check step.
Ian.
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them
2015-03-24 12:40 ` Ian Jackson
@ 2015-03-24 12:46 ` Ian Campbell
2015-03-24 18:35 ` Ian Jackson
0 siblings, 1 reply; 10+ messages in thread
From: Ian Campbell @ 2015-03-24 12:46 UTC (permalink / raw)
To: Ian Jackson; +Cc: xen-devel
On Tue, 2015-03-24 at 12:40 +0000, Ian Jackson wrote:
> Ian Campbell writes ("[PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them"):
> > Refactor the $kvp_replace helper in ts-xen-install into a generic
> > helper (which requires using ::EO and ::EI for namespacing) for use
> > with target_editfile [etc.]
>
> Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
>
> > I did debate making the presence of cores in /var/core fail the test
> > (somehow), but I decided that would be annoying for standalone mode or
> > shared host scenarios where the core files might be stale or related
> > to another job.
>
> I think you should add /var/core to the leak check step.
Which in practice just means listing it in the arugment to
inventory_files?
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them
2015-03-24 12:46 ` Ian Campbell
@ 2015-03-24 18:35 ` Ian Jackson
2015-03-25 9:40 ` Ian Campbell
0 siblings, 1 reply; 10+ messages in thread
From: Ian Jackson @ 2015-03-24 18:35 UTC (permalink / raw)
To: Ian Campbell; +Cc: xen-devel
Ian Campbell writes ("Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them"):
> On Tue, 2015-03-24 at 12:40 +0000, Ian Jackson wrote:
> > I think you should add /var/core to the leak check step.
>
> Which in practice just means listing it in the arugment to
> inventory_files?
Yes.
Ian.
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them
2015-03-24 18:35 ` Ian Jackson
@ 2015-03-25 9:40 ` Ian Campbell
2015-03-25 14:38 ` Ian Campbell
2015-03-25 16:18 ` Ian Jackson
0 siblings, 2 replies; 10+ messages in thread
From: Ian Campbell @ 2015-03-25 9:40 UTC (permalink / raw)
To: Ian Jackson; +Cc: xen-devel
On Tue, 2015-03-24 at 18:35 +0000, Ian Jackson wrote:
> Ian Campbell writes ("Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them"):
> > On Tue, 2015-03-24 at 12:40 +0000, Ian Jackson wrote:
> > > I think you should add /var/core to the leak check step.
> >
> > Which in practice just means listing it in the arugment to
> > inventory_files?
>
> Yes.
Good, v2 in <1427207038-28090-1-git-send-email-ian.campbell@citrix.com>
does just that.
Do you have any thoughts on how to best universally arrange for the
rlimits to be increased? Probably inserting a ulimit call into the
libvirt initscript would be an easy first step, and since that's the
thing we most often see crashing it seems. I'll send such a patch
shortly.
Adding it to xencommons would be OK too, which just leaves catching e.g.
daemonised xl processes launched via ssh. Hacking /etc/profile or
something in TestSupport::cmdex perhaps?
Ian.
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them
2015-03-25 9:40 ` Ian Campbell
@ 2015-03-25 14:38 ` Ian Campbell
2015-03-25 16:18 ` Ian Jackson
1 sibling, 0 replies; 10+ messages in thread
From: Ian Campbell @ 2015-03-25 14:38 UTC (permalink / raw)
To: Ian Jackson; +Cc: xen-devel
On Wed, 2015-03-25 at 09:40 +0000, Ian Campbell wrote:
> Do you have any thoughts on how to best universally arrange for the
> rlimits to be increased? Probably inserting a ulimit call into the
> libvirt initscript would be an easy first step, and since that's the
> thing we most often see crashing it seems. I'll send such a patch
> shortly.
>
> Adding it to xencommons would be OK too, which just leaves catching e.g.
> daemonised xl processes launched via ssh. Hacking /etc/profile or
> something in TestSupport::cmdex perhaps?
I handled this by writing an /etc/security/limits.d/ file during host
install, which seems to have done the trick. I just sent a patch which
along with the libvirt one I sent earlier today covers everything I
think we care about except our daemons which are launched from
xencommons.
The initscript in xen.git doesn't offer a useful hook for this, we could
add something to the end of /etc/default/xencommons, which is sourced by
the script, but that seems like a bit of an abuse.
Ian.
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them
2015-03-25 9:40 ` Ian Campbell
2015-03-25 14:38 ` Ian Campbell
@ 2015-03-25 16:18 ` Ian Jackson
2015-03-25 16:30 ` Ian Campbell
1 sibling, 1 reply; 10+ messages in thread
From: Ian Jackson @ 2015-03-25 16:18 UTC (permalink / raw)
To: Ian Campbell; +Cc: xen-devel
Ian Campbell writes ("Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them"):
> Do you have any thoughts on how to best universally arrange for the
> rlimits to be increased? Probably inserting a ulimit call into the
> libvirt initscript would be an easy first step, and since that's the
> thing we most often see crashing it seems. I'll send such a patch
> shortly.
Maybe pam ?
Ian.
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them
2015-03-25 16:18 ` Ian Jackson
@ 2015-03-25 16:30 ` Ian Campbell
0 siblings, 0 replies; 10+ messages in thread
From: Ian Campbell @ 2015-03-25 16:30 UTC (permalink / raw)
To: Ian Jackson; +Cc: xen-devel
On Wed, 2015-03-25 at 16:18 +0000, Ian Jackson wrote:
> Ian Campbell writes ("Re: [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them"):
> > Do you have any thoughts on how to best universally arrange for the
> > rlimits to be increased? Probably inserting a ulimit call into the
> > libvirt initscript would be an easy first step, and since that's the
> > thing we most often see crashing it seems. I'll send such a patch
> > shortly.
>
> Maybe pam ?
Yes, I think that's what my subsequent patch to configure
via /etc/security/limits.d has achived.
Ian.
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2015-03-25 16:30 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-03-09 15:40 [PATCH OSSTEST] Arrange for core dumps to be placed in /var/core and collect them Ian Campbell
2015-03-09 21:54 ` Don Slutz
2015-03-10 9:38 ` Ian Campbell
2015-03-24 12:40 ` Ian Jackson
2015-03-24 12:46 ` Ian Campbell
2015-03-24 18:35 ` Ian Jackson
2015-03-25 9:40 ` Ian Campbell
2015-03-25 14:38 ` Ian Campbell
2015-03-25 16:18 ` Ian Jackson
2015-03-25 16:30 ` Ian Campbell
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.