Re: 6.3 just died (not for the first time)

Hi Martin and all others involved, On 5/16/18 10:20 AM, Martin Pieuchot wrote: On 16/05/18(Wed) 08:06, Harald Dunkel wrote: Hi folks, Thanks for the report. I highly appreciate the syspatch. Thanx very much Regards Harri

Re: 6.3 just died (not for the first time)

On 6/26/18 10:56 AM, Harald Dunkel wrote: Hi folks, Is it possible that this problem was introduced by syspatch 008_ipsecout ? AFAICS my hosts without 008_ipsecout did not die, so I take the silence as a "yes". Regards Harri

Re: 6.3 just died (not for the first time)

Hi folks, On 5/16/18 10:20 AM, Martin Pieuchot wrote: On 16/05/18(Wed) 08:06, Harald Dunkel wrote: Hi folks, Thanks for the report. hopefully its allowed to repost this message here: One gateway running 6.3 ran into the debugger last night. Last words: login: kernel: protection fault

Re: 6.3 just died (not for the first time)

Mon, 28 May 2018 20:30:33 +0200 Harald Dunkel > On 5/25/18 3:20 PM, Alexander Bluhm wrote: > > On Tue, May 22, 2018 at 07:32:29AM +0200, Harald Dunkel wrote: > >> Do you think this is worth a syspatch? > > > > You are asking this question for three times. The answer is no. > > > >

Re: 6.3 just died (not for the first time)

On 5/25/18 3:20 PM, Alexander Bluhm wrote: > On Tue, May 22, 2018 at 07:32:29AM +0200, Harald Dunkel wrote: >> Do you think this is worth a syspatch? > > You are asking this question for three times. The answer is no. > Twice, IIRC. My apologies. Thanx for your answer. I am glad that I could

Re: 6.3 just died (not for the first time)

On Fri, May 25, 2018 at 03:20:50PM +0200, Alexander Bluhm wrote: > On Tue, May 22, 2018 at 07:32:29AM +0200, Harald Dunkel wrote: > > Do you think this is worth a syspatch? > > You are asking this question for three times. The answer is no. > > Errata and syspatch are for security issues that

Re: 6.3 just died (not for the first time)

On Tue, May 22, 2018 at 07:32:29AM +0200, Harald Dunkel wrote: > Do you think this is worth a syspatch? You are asking this question for three times. The answer is no. Errata and syspatch are for security issues that can be triggered remotely or by a local non-root user. If we would fix every

Re: 6.3 just died (not for the first time)

Hi folks, On 5/16/18 1:07 PM, Stuart Henderson wrote: If you can get it tested as a patch on a self-built -stable kernel and we can get it committed then there's a *chance* it might be errata worthy in which case you could go back to syspatches instead. The patched kernel is running for a

Re: 6.3 just died (not for the first time)

Hi folks, On 5/16/18 3:32 PM, Harald Dunkel wrote: On 5/16/18 10:20 AM, Martin Pieuchot wrote: That means that the TDB has already been freed.  This is possible because the timeout sleeps on the NET_LOCK().  Diff below should prevent that by introducing a tdb_reaper() function like we do in

Re: 6.3 just died (not for the first time)

On Wed, May 16, 2018 at 10:20:49AM +0200, Martin Pieuchot wrote: > That means that the TDB has already been freed. This is possible > because the timeout sleeps on the NET_LOCK(). Diff below should prevent > that by introducing a tdb_reaper() function like we do in other parts of > the stack.

Re: 6.3 just died (not for the first time)

On 2018/05/16 13:37, Harald Dunkel wrote: > On 5/16/18 1:07 PM, Stuart Henderson wrote: > > > > You can't create a syspatch from it. > > I am talking about a *private* syspatch. Something that I can > revert later to gain access to the official syspatches again. > That should be possible. Oh.

Re: 6.3 just died (not for the first time)

On 5/16/18 1:07 PM, Stuart Henderson wrote: You can't create a syspatch from it. I am talking about a *private* syspatch. Something that I can revert later to gain access to the official syspatches again. That should be possible. If you can get it tested as a patch on a self-built -stable

Re: 6.3 just died (not for the first time)

On Wed, May 16, 2018 at 12:34:57PM +0200, Harald Dunkel wrote: > Thanx for the patch, but I wonder how I can create a syspatch from > it? You can't. The syspatch utility only installs officially signed syspatches and you don't have the signing key. > If I patch, build and install stable from

Re: 6.3 just died (not for the first time)

On 2018/05/16 12:34, Harald Dunkel wrote: > Thanx for the patch, but I wonder how I can create a syspatch from > it? If I patch, build and install stable from source, then my hosts > are cut off from the syspatch scheme. That would be highly painful. > > Every helpful comment is highly

Re: 6.3 just died (not for the first time)

On Wed, 16 May 2018 12:34:57 +0200 > Thanx for the patch, but I wonder how I can create a syspatch from > it? If I patch, build and install stable from source, then my hosts > are cut off from the syspatch scheme. That would be highly painful. > > Every helpful comment is highly appreciated.

Re: 6.3 just died (not for the first time)

Thanx for the patch, but I wonder how I can create a syspatch from it? If I patch, build and install stable from source, then my hosts are cut off from the syspatch scheme. That would be highly painful. Every helpful comment is highly appreciated. Harri

Re: 6.3 just died (not for the first time)

On 16/05/18(Wed) 08:06, Harald Dunkel wrote: > Hi folks, Thanks for the report. > hopefully its allowed to repost this message here: > > One gateway running 6.3 ran into the debugger last night. Last words: > > login: kernel: protection fault trap, code=0 > Stopped at export_sa+0x5c:

6.3 just died (not for the first time)

Hi folks, hopefully its allowed to repost this message here: One gateway running 6.3 ran into the debugger last night. Last words: login: kernel: protection fault trap, code=0 Stopped at export_sa+0x5c: movl0(%rcx),%ecx ddb{0}> show panic the kernel did not panic ddb{0}> trace