On Wed, Feb 08, 2023 at 11:37:11AM +0100, Thomas Pedoussaut wrote:
>
> On 08/02/2023 10:09, Aurelien DARRAGON wrote:
> > > In fact at some point I had a backend with 5 srv from config + 3
> > > dynamically inserted. Those new ones got about 50 requests pushed to
> > > them, until they reaches the
On 08/02/2023 10:09, Aurelien DARRAGON wrote:
In fact at some point I had a backend with 5 srv from config + 3
dynamically inserted. Those new ones got about 50 requests pushed to
them, until they reaches the slowstart delay(I think, must investigate
more), when they stopped being selected for
On Wed, Feb 08, 2023 at 10:18:07AM +0100, Willy Tarreau wrote:
> On Wed, Feb 08, 2023 at 09:58:47AM +0100, Thomas Pedoussaut wrote:
> > On 08/02/2023 09:52, Willy Tarreau wrote:
> > > Just out of curiosity (and in order to help narrow the root cause
> > > further),
> > > it would be interesting to
On Wed, Feb 08, 2023 at 09:58:47AM +0100, Thomas Pedoussaut wrote:
> On 08/02/2023 09:52, Willy Tarreau wrote:
> > Just out of curiosity (and in order to help narrow the root cause
> > further),
> > it would be interesting to know if the same problem happens with static
> > algorithms like "balance
> In fact at some point I had a backend with 5 srv from config + 3
> dynamically inserted. Those new ones got about 50 requests pushed to
> them, until they reaches the slowstart delay(I think, must investigate
> more), when they stopped being selected for new connection. The status
> was L7OK for
On 08/02/2023 09:52, Willy Tarreau wrote:
Just out of curiosity (and in order to help narrow the root cause
further),
it would be interesting to know if the same problem happens with static
algorithms like "balance uri", as they cannot use the same heuristics as
other algos like "random" or "rou
On Wed, Feb 08, 2023 at 09:46:52AM +0100, Thomas Pedoussaut wrote:
> Thanks Aurelien, I feel less lonely on this.
:-)
> I might add that sometime as well the server check results switches to 4 =
> CHK_RES_CONDPASS which seems to indicate a bug in the handling of the
> maxconn parameter.
>
> I wi
Thanks Aurelien, I feel less lonely on this.
I might add that sometime as well the server check results switches to 4
= CHK_RES_CONDPASS which seems to indicate a bug in the handling of the
maxconn parameter.
I will run experiments without the maxconn to confirm (or not) the culprit.
On 08/0
On Wed, Feb 08, 2023 at 09:38:41AM +0100, Aurelien DARRAGON wrote:
> Hi,
>
> I don't know if it could help, but based on Thomas instructions/example,
> I'm able to reproduce:
>
> As weird as it may seem, the 'maxconn' parameter used with dynamic
> server seems to trigger the issue.
>
> Below is
Hi,
I don't know if it could help, but based on Thomas instructions/example,
I'm able to reproduce:
As weird as it may seem, the 'maxconn' parameter used with dynamic
server seems to trigger the issue.
Below is a minimal reproducer:
> global
> stats socket /tmp/ha.sock mode 660 level admin e
10 matches
Mail list logo
