Hi!
I'd like to raise the question of FPM SAPI maintainership. Is it still
maintained? If so, by whom? Jerome Loyet is listed as a maintainer, but
his last commit has been in 2012, as far as I can see, and I see no
activity from him on the bug tracker for the last couple of years (maybe
more). FPM is a pretty important module and widely used, and we have
over 100 bugs in the bug DB for it, not being addressed. Would anyone
like to step up for his module? If not, what's the plan for keeping it
alive and well?
Thanks,
Stas Malyshev
smalyshev@gmail.com
Hi!
I'd like to raise the question of FPM SAPI maintainership. Is it still
maintained? If so, by whom? Jerome Loyet is listed as a maintainer, but
his last commit has been in 2012, as far as I can see, and I see no
activity from him on the bug tracker for the last couple of years (maybe
more). FPM is a pretty important module and widely used, and we have
over 100 bugs in the bug DB for it, not being addressed. Would anyone
like to step up for his module? If not, what's the plan for keeping it
alive and well?
I'd like to second that. Given the past development of heading away from
SAPIs loaded into the webserver to FastCGI being preferred, we should be
giving FPM the deserved priority. Does anyone have stats about the
popularity of our SAPIs?
I also have a patch I'd like to discuss with the new maintainers, which
attempts to fix FPM's SIGUSR2 graceful restart behaviour, and which
we've been running for half a year now IIRC.
--
Regards,
Mike
Hi,
Le 21/02/2018 à 10:21, Michael Wallner a écrit :
I'd like to second that. Given the past development of heading away from
SAPIs loaded into the webserver to FastCGI being preferred,
+1
we should be
giving FPM the deserved priority. Does anyone have stats about the
popularity of our SAPIs?
I don't really have statistic, but Fedora use FPM by default (mod_php is
available but disabled) since F27 (mostly because httpd have switched to
event mode by default, which is now required for mod_http2).
I also have a patch I'd like to discuss with the new maintainers, which
attempts to fix FPM's SIGUSR2 graceful restart behaviour, and which
we've been running for half a year now IIRC.
Any link to the patch ?
Remi
Hi,
Le 21/02/2018 à 10:21, Michael Wallner a écrit :
I also have a patch I'd like to discuss with the new maintainers, which
attempts to fix FPM's SIGUSR2 graceful restart behaviour, and which
we've been running for half a year now IIRC.Any link to the patch ?
Sure, attached!
--
Regards,
Mike
Hi!
How to get this going forward?
Hi,
Le 21/02/2018 à 10:21, Michael Wallner a écrit :
I also have a patch I'd like to discuss with the new maintainers, which
attempts to fix FPM's SIGUSR2 graceful restart behaviour, and which
we've been running for half a year now IIRC.Any link to the patch ?
Sure, attached!
--
Regards,
Mike
Hi!
How to get this going forward?
Hi,
Le 21/02/2018 à 10:21, Michael Wallner a écrit :
I also have a patch I'd like to discuss with the new maintainers, which
attempts to fix FPM's SIGUSR2 graceful restart behaviour, and which
we've been running for half a year now IIRC.Any link to the patch ?
Sure, attached!
Please submit patches on GitHub.
Nikita
Hey
On Wed, Feb 21, 2018 at 12:42 AM, Stanislav Malyshev smalyshev@gmail.com
wrote:
Hi!
I'd like to raise the question of FPM SAPI maintainership. Is it still
maintained? If so, by whom? Jerome Loyet is listed as a maintainer, but
his last commit has been in 2012, as far as I can see, and I see no
activity from him on the bug tracker for the last couple of years (maybe
more). FPM is a pretty important module and widely used, and we have
over 100 bugs in the bug DB for it, not being addressed. Would anyone
like to step up for his module? If not, what's the plan for keeping it
alive and well?
I actually work on some parts of FPM and trying to help with some PR's
already. Currently I work on the rewrite of zlog and as part of that I
completely rewrote the test framework which is inconvenient for some
scenarios (especially logging but there are other parts that the current
tests are not good for). All work is currently done on my FPMi project repo
[1] which is meant to be a dev version of FPM and potentially separated
SAPI that can bring the next features sooner to users on the older PHP
versions. I also started working to the tool for better testing some time
ago [2]. It's currently just a simple fastcgi client but plan to resume
work on it and extend it so it can be used for regression testing and
recreating some bugs that are not possible to do in PHP tests.
The zlog part is more or less working but needs optimization (better
buffering) and some final improvements and fixes. That's probably most
wanted feature especially for Docker execution. Then I would like to use
the new testing framework for all tests and look to other issues.
I already work on it and not sure if there is any other dev actively works
on it so I would be happy to step up as a maintainer whatever it means :) I
want however keep working on my plan so the current priority is to sort the
logging out because it is causing many issues.
[1] https://github.com/bukka/fpmi/compare/0.1.0...next
[2] https://github.com/bukka/fpmt
Cheers
Jakub
Yeah I was going to say: work with Jakub, he has some good stuff in flight! :)
Hey
On Wed, Feb 21, 2018 at 12:42 AM, Stanislav Malyshev smalyshev@gmail.com
wrote:Hi!
I'd like to raise the question of FPM SAPI maintainership. Is it still
maintained? If so, by whom? Jerome Loyet is listed as a maintainer, but
his last commit has been in 2012, as far as I can see, and I see no
activity from him on the bug tracker for the last couple of years (maybe
more). FPM is a pretty important module and widely used, and we have
over 100 bugs in the bug DB for it, not being addressed. Would anyone
like to step up for his module? If not, what's the plan for keeping it
alive and well?I actually work on some parts of FPM and trying to help with some PR's
already. Currently I work on the rewrite of zlog and as part of that I
completely rewrote the test framework which is inconvenient for some
scenarios (especially logging but there are other parts that the current
tests are not good for). All work is currently done on my FPMi project repo
[1] which is meant to be a dev version of FPM and potentially separated
SAPI that can bring the next features sooner to users on the older PHP
versions. I also started working to the tool for better testing some time
ago [2]. It's currently just a simple fastcgi client but plan to resume
work on it and extend it so it can be used for regression testing and
recreating some bugs that are not possible to do in PHP tests.The zlog part is more or less working but needs optimization (better
buffering) and some final improvements and fixes. That's probably most
wanted feature especially for Docker execution. Then I would like to use
the new testing framework for all tests and look to other issues.I already work on it and not sure if there is any other dev actively works
on it so I would be happy to step up as a maintainer whatever it means :) I
want however keep working on my plan so the current priority is to sort the
logging out because it is causing many issues.[1] https://github.com/bukka/fpmi/compare/0.1.0...next
[2] https://github.com/bukka/fpmtCheers
Jakub
Hi!
I already work on it and not sure if there is any other dev actively
works on it so I would be happy to step up as a maintainer whatever it
means :) I want however keep working on my plan so the current priority
is to sort the logging out because it is causing many issues.
Great! There are also a number of issues in the bug db for FPM that
could use a review, and I am not sure what to do about
https://bugs.php.net/bug.php?id=75605 specifically...
--
Stas Malyshev
smalyshev@gmail.com
Hi!
I already work on it and not sure if there is any other dev actively
works on it so I would be happy to step up as a maintainer whatever it
means :) I want however keep working on my plan so the current priority
is to sort the logging out because it is causing many issues.
Great! There are also a number of issues in the bug db for FPM that
could use a review, and I am not sure what to do about
https://bugs.php.net/bug.php?id=75605 specifically...
--
Stas Malyshev
smalyshev@gmail.com
Great! There are also a number of issues in the bug db for FPM that
could use a review, and I am not sure what to do about
https://bugs.php.net/bug.php?id=75605 specifically...
You do realize that this bug is marked private, thus nobody can say
anything about it...
--
regards Helmut K. C. Tessarek KeyID 0x172380A011EF4944
Key fingerprint = 8A55 70C1 BD85 D34E ADBC 386C 1723 80A0 11EF 4944
/*
Thou shalt not follow the NULL
pointer for chaos and madness
await thee at its end.
*/
Hi!
Great! There are also a number of issues in the bug db for FPM that
could use a review, and I am not sure what to do about
https://bugs.php.net/bug.php?id=75605 specifically...You do realize that this bug is marked private, thus nobody can say
anything about it...
Not nobody - Jakub actually does have access to it :)
--
Stas Malyshev
smalyshev@gmail.com
Hi Stas,
You do realize that this bug is marked private, thus nobody can say
anything about it...Not nobody - Jakub actually does have access to it :)
Ok, because I've even tried with my php id and I could still not see it.
Anyway, I'd be interested in helping out with FPM as well.
Jakub, it looks like you are going to be the new lead, so please let me
know, if you want me to look into specific issues.
Cheers,
K. C.
--
regards Helmut K. C. Tessarek KeyID 0x172380A011EF4944
Key fingerprint = 8A55 70C1 BD85 D34E ADBC 386C 1723 80A0 11EF 4944
/*
Thou shalt not follow the NULL
pointer for chaos and madness
await thee at its end.
*/