[libvirt] [PATCH] util: ignore SIGPIPE when writing to stderr/stdout

[Daniel P. Berrangé]

On Sat, Oct 12, 2019 at 04:04:11PM +0800, Wang Yechao wrote:
> libvirtd's stderr/stdout redirected to journald by default, so if
> the journald has stopped, libvirtd and libvirtd's child process
> will receive SIGPIPE signal when writing logs to stderr/stdout.
> 
> journald stopped reasons:
> 1. manual command "systemctl stop systemd-journald.service"

Don't do that then IMHO.

> 2. oom killer kill it.
> ...

If OOM killer picked journald, then it is game over for the host IMHO
and thed right answer is a reboot. This would be very surprising though.
OOM killer is more likely to pick QEMU or libvirtd as they're bigger
targets.

> 
> Signed-off-by: Wang Yechao <wang.yechao255 at zte.com.cn>
> ---
>  src/util/virlog.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/src/util/virlog.c b/src/util/virlog.c
> index 4c76fbc..127e121 100644
> --- a/src/util/virlog.c
> +++ b/src/util/virlog.c
> @@ -39,6 +39,7 @@
>  # include <sys/un.h>
>  #endif
>  #include <fnmatch.h>
> +#include <signal.h>
>  
>  #include "virerror.h"
>  #include "virlog.h"
> @@ -732,6 +733,9 @@ virLogOutputToFd(virLogSourcePtr source ATTRIBUTE_UNUSED,
>      if (fd < 0)
>          return;
>  
> +    if (fd == STDERR_FILENO || fd == STDOUT_FILENO)
> +        signal(SIGPIPE, SIG_IGN);

This is very wrong....

This code is run in both client & server, so we're messing with SIGPIPE
in applications that link to libvirt.so.

The use of signal() is not safe in multi-threaded applications.

In libvirtd we have set SIGPIPE to ignored already when starting up
libvirtd.


Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|