|
OpenBSD Packages and Ports Installation and upgrading of packages and ports on OpenBSD. |
|
Thread Tools | Display Modes |
|
||||
I went back and rebuilt the httpd module from -current source, and now cannot reproduce the problem on -current. I saw it with a recent snapshot, but not with httpd rebuilt from -current source.
I'll spend some time on this over the weekend looking at commits between -release and -current, and see if I can determine which commit resolved the problem. |
|
|||
Thanks, jggimi. I'm afraid I don't think I'd be much help parsing the source code but, nonetheless, would be happy to do so with you if you like?
I was going to say, I'd hazard a guess that it's a bug in the operation not the documentation. Also, I'm surprised that this hasn't been picked up earlier. Could it be that it's a recently introduced bug; that is, up until a recent commit it did function as expected? Because surely someone ran into real problems in production with this fault? It's a pretty significant service disruption. |
|
||||
At this time, it looks to me as if:
|
|
||||
I have retested in the -current snapshot, with the snapshot's binary of /usr/sbin/httpd, and could not reproduce the problem.
It appears, now, that this has been fixed in -current. I do not know why my first test in -current appeared to reproduce the problem. |
|
|||
You've done a lot already. It's an interesting find. I mean, it looks like it previously worked as the documentation suggests it should, then broke, but hadn't been picked up till now. I'm looking forward to not only seeing it fixed but understanding why it broke.
|
|
||||
It's not clear to me, yet, under what conditions it fails. I planned to begin trying to isolate those conditions when it would fail or not fail, but I won't have time to start until next week.
I tripped over my own assumptions regarding DEBUG settings. And I haven't had enough time to research this completely. I have not walked through the code yet, only noted in which modules the variable maxrequestbody gets set, and where it gets tested. I hope, next week, to study the conditions under which the bug appears, and step through the code line by line, and compare that with the code path taken in conditions where the bug does not manifest. I'm not a developer, and have only just begun delving into this program. |
|
|||
You've done well getting this far already. Is the developer aware of this bug yet?
|
Tags |
httpd, mariadb, mysql, nextcloud, php |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
httpd rc.conf.local question | psypro | OpenBSD General | 3 | 30th October 2016 05:54 PM |
httpd.conf chroot | morophla | OpenBSD General | 4 | 19th April 2015 02:07 PM |
ipsec.conf & IKE mode config | pyzo | OpenBSD Security | 1 | 24th May 2012 05:06 AM |
Update httpd.conf IPs from DNS zones. | bigb89 | Programming | 16 | 2nd December 2008 02:02 AM |
httpd.conf | Snoop1990 | General software and network | 5 | 29th July 2008 04:30 AM |