[Bug 263062] tcp_inpcb leaking in VM environment

2023-11-03 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 Eugene Grosbein changed: What|Removed |Added Resolution|--- |FIXED Status|In Prog

[Bug 263062] tcp_inpcb leaking in VM environment

2023-10-16 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 Eugene Grosbein changed: What|Removed |Added Status|Open|In Progress --- Comment #17 from

[Bug 263062] tcp_inpcb leaking in VM environment

2023-10-16 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #16 from Max Khon --- Yes, I think this can be closed now -- You are receiving this mail because: You are the assignee for the bug.

[Bug 263062] tcp_inpcb leaking in VM environment

2023-10-16 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 John Baldwin changed: What|Removed |Added CC||j...@freebsd.org --- Comment #15 fr

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-30 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 paul vixie changed: What|Removed |Added CC||p...@redbarn.org --- Comment #14 from

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-30 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 t...@ida2.de changed: What|Removed |Added CC||t...@ida2.de --- Comment #13 from t.

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-10 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #12 from Gleb Smirnoff --- I'm not 100% sure D38898 resolves all of inpcb problems on stable/13, as the patch is irrelevant to inpcbs. However, it definitely is a requirement for proper operation in some VM environments. -- Yo

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-10 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 Max Khon changed: What|Removed |Added See Also||https://bugs.freebsd.org/bu

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-10 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #11 from Max Khon --- With the help from glebius@, kib@ and afedorov@ the root cause has been identified to be this: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269823 The patch (https://reviews.freebsd.org/D38898) resolv

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-09 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #10 from Max Khon --- Eugene, if the problem is still actual for you, this is how you can get the chipset: --- cut here --- $ sudo dmidecode -s baseboard-product-name Standard PC (Q35 + ICH9, 2009) $ --- cut here --- On i440fx

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-09 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #9 from Max Khon --- I will try again with GENERIC-NODEBUG kernel -- You are receiving this mail because: You are the assignee for the bug.

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-09 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #8 from Max Khon --- I built 14.0-CURRENT GENERIC kernel but "ab -c 100" makes nginx worker dormant (stuck, not answering and in unkillable state) faster than it was on 13.2-RELEASE. So I don't get to the point where it may star

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-07 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #7 from Gleb Smirnoff --- Max, I'd appreciate if you run same tests on 14.0-CURRENT. The inpcb subsystem is significantly rewritten there. If the bug is still there, that means it is outside of inpcb engine, but is isolated to o

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-07 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #6 from Max Khon --- I can confirm that switching Hetzner VM to i440fx (rescale to Intel plan, then ask Hetzner support to switch to i440fx as some Intel VMs are also provisioned with Q35 chipset) solves the issue (on the same 1

[Bug 263062] tcp_inpcb leaking in VM environment

2023-08-04 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 Max Khon changed: What|Removed |Added CC||f...@freebsd.org --- Comment #5 from Ma

[Bug 263062] tcp_inpcb leaking in VM environment

2022-09-24 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #4 from Eugene M. Zheganin --- Also, I found a post on the FreeBSD forums where it is said that this exhaustion is probably only happening to a FreeBSD running on a modern Q35 chipset, same public cloud Hetzner mentioned (my VMs

[Bug 263062] tcp_inpcb leaking in VM environment

2022-09-24 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #3 from Eugene M. Zheganin --- Follow up: the latter VM is running net-mgmt/prometheus2 and sysutils/snmp-exporter only, so what at first look may have appeared as being triggered by some self-developed application issue may hav

[Bug 263062] tcp_inpcb leaking in VM environment

2022-09-24 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 --- Comment #2 from Eugene M. Zheganin --- Same stuff happends with udp_inpcb: [zone: udp_inpcb] kern.ipc.maxsockets limit reached [zone: udp_inpcb] kern.ipc.maxsockets limit reached [zone: udp_inpcb] kern.ipc.maxsockets limit reached [zon

[Bug 263062] tcp_inpcb leaking in VM environment

2022-04-06 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263062 Mark Linimon changed: What|Removed |Added Assignee|b...@freebsd.org|virtualization@FreeBSD.org -- You