[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#980248: marked as done (cgit: kernel regression breaks cgit's sendfile())

Your message dated Mon, 8 Feb 2021 21:07:00 -0500
with message-id <YCHuRCLMO7oQqmK6@colberg.org>
and subject line Re: Bug#980248: cgit: kernel regression breaks cgit's sendfile()
has caused the Debian Bug report #980248,
regarding cgit: kernel regression breaks cgit's sendfile()
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
980248: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=980248
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: cgit
Version: 1.2.3+git2.25.1-1
Severity: normal

Dear Maintainer,

Having changed my virtual machine setup to run kernel 5.10.4,
I've noticed that a kernel regression breaks cgit's use of
sendfile() in e.g. fcgiwrap setups like mine, because sendfile()
to a pipe() no longer works. See also

https://lore.kernel.org/lkml/C8KER7U60WXE.25UFD8RE6QZQK@oguc/

(I reported the additional pipe() issue there, bug the list is
currently down). A simple test program I wrote is here:

https://p.sipsolutions.net/b8b30d96fd9428a8.txt


In any case - this is obviously not really a bug in cgit but
rather in the kernel, but I thought perhaps a warning might be
something you'd appreciate.

-- System Information:
Debian Release: bullseye/sid
  APT prefers testing
  APT policy: (990, 'testing'), (500, 'unstable'), (500, 'stable'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 5.10.4+ (SMP w/6 CPU threads)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /bin/bash
Init: systemd (via /run/systemd/system)

Versions of packages cgit depends on:
ii  libc6        2.31-6
ii  liblua5.1-0  5.1.5-8.1+b3
ii  zlib1g       1:1.2.11.dfsg-2

Versions of packages cgit recommends:
ii  nginx-core [httpd]  1.18.0-6+b1
ii  nginx-full [httpd]  1.18.0-6

Versions of packages cgit suggests:
ii  python3           3.9.1-1
ii  python3-docutils  0.16+dfsg-3
pn  python3-markdown  <none>
ii  python3-pygments  2.7.1+dfsg-1

-- Configuration Files:
/etc/cgitrc changed [not included]

-- no debconf information

--- End Message ---
--- Begin Message --- 
Hi Johannes,

On Sat, Jan 16, 2021 at 05:54:43PM +0100, Johannes Berg wrote:
> Having changed my virtual machine setup to run kernel 5.10.4,
> I've noticed that a kernel regression breaks cgit's use of
> sendfile() in e.g. fcgiwrap setups like mine, because sendfile()
> to a pipe() no longer works. See also
> 
> https://lore.kernel.org/lkml/C8KER7U60WXE.25UFD8RE6QZQK@oguc/
> 
> (I reported the additional pipe() issue there, bug the list is
> currently down). A simple test program I wrote is here:
> 
> https://p.sipsolutions.net/b8b30d96fd9428a8.txt

Thank you for heads-up. I checked as we are getting close to the soft
freeze for bullseye, and the issue appears resolved in linux 5.10.11:

https://bugzilla.kernel.org/show_bug.cgi?id=211217#c2

Regards,
Peter

--- End Message ---
Reply to: