Re: fqcodel panic

2017-05-25 Thread Mike Belopuhov
On Thu, May 25, 2017 at 14:54 +0200, Sebastien Marie wrote: > On Thu, May 25, 2017 at 01:14:54PM +0200, Mike Belopuhov wrote: > > > > > > Sorry I didn't test it: I had understood that due the problem was due to > > > my (unsupported) use of fq on pppoe0, so I just skipped it (the > > > backtrace o

Re: fqcodel panic

2017-05-25 Thread Sebastien Marie
On Thu, May 25, 2017 at 01:14:54PM +0200, Mike Belopuhov wrote: > > > > Sorry I didn't test it: I had understood that due the problem was due to > > my (unsupported) use of fq on pppoe0, so I just skipped it (the > > backtrace of the panic passed by sppp_ functions). > > > > We still need to solv

Re: fqcodel panic

2017-05-25 Thread Mike Belopuhov
On Thu, May 25, 2017 at 07:25 +0200, Sebastien Marie wrote: > On Wed, May 24, 2017 at 06:03:04PM +0200, Mike Belopuhov wrote: > > On Sun, May 21, 2017 at 11:53 +0200, Mike Belopuhov wrote: > > > On Sat, May 20, 2017 at 18:53 +0200, Mike Belopuhov wrote: > > > > On Sat, May 20, 2017 at 17:30 +0200,

Re: fqcodel panic

2017-05-24 Thread Sebastien Marie
On Wed, May 24, 2017 at 06:03:04PM +0200, Mike Belopuhov wrote: > On Sun, May 21, 2017 at 11:53 +0200, Mike Belopuhov wrote: > > On Sat, May 20, 2017 at 18:53 +0200, Mike Belopuhov wrote: > > > On Sat, May 20, 2017 at 17:30 +0200, Sebastien Marie wrote: > > > > > > > > From my pf configuration, I

Re: fqcodel panic - pfctl

2017-05-24 Thread Mike Belopuhov
On Sat, May 20, 2017 at 18:03 +0200, Sebastien Marie wrote: > Hi, > > Here a report for a second panic regarding fqcodel. > > The panic occurs at pfctl time: KASSERT(qs->parent_qid == 0) from > fqcodel_pf_addqueue is reached. > > I suspect a fqcodel queue not to be usable as parent of another qu

Re: fqcodel panic

2017-05-24 Thread Mike Belopuhov
On Sun, May 21, 2017 at 11:53 +0200, Mike Belopuhov wrote: > On Sat, May 20, 2017 at 18:53 +0200, Mike Belopuhov wrote: > > On Sat, May 20, 2017 at 17:30 +0200, Sebastien Marie wrote: > > > Hi, > > > > > > Yesterday, I upgraded my amd64 gateway (pppoe + vlans) with a recent > > > -current in order

Re: fqcodel panic

2017-05-21 Thread Mike Belopuhov
On Sat, May 20, 2017 at 18:53 +0200, Mike Belopuhov wrote: > On Sat, May 20, 2017 at 17:30 +0200, Sebastien Marie wrote: > > Hi, > > > > Yesterday, I upgraded my amd64 gateway (pppoe + vlans) with a recent > > -current in order to test a bit codel queue. > > > > OpenBSD 6.1-current (GENERIC.MP) #

Re: fqcodel panic

2017-05-20 Thread Mike Belopuhov
On Sat, May 20, 2017 at 17:30 +0200, Sebastien Marie wrote: > Hi, > > Yesterday, I upgraded my amd64 gateway (pppoe + vlans) with a recent > -current in order to test a bit codel queue. > > OpenBSD 6.1-current (GENERIC.MP) #67: Thu May 18 18:28:26 MDT 2017 > dera...@amd64.openbsd.org:/usr/src

fqcodel panic - pfctl

2017-05-20 Thread Sebastien Marie
Hi, Here a report for a second panic regarding fqcodel. The panic occurs at pfctl time: KASSERT(qs->parent_qid == 0) from fqcodel_pf_addqueue is reached. I suspect a fqcodel queue not to be usable as parent of another queue. but pfctl parser doesn't forbid it, and kernel panic occurs. $ cat pf

fqcodel panic

2017-05-20 Thread Sebastien Marie
Hi, Yesterday, I upgraded my amd64 gateway (pppoe + vlans) with a recent -current in order to test a bit codel queue. OpenBSD 6.1-current (GENERIC.MP) #67: Thu May 18 18:28:26 MDT 2017 dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP This first panic occured while runnin