[squid-users] Fw: [Bug 4977] stmem.cc:98: "lowestOffset () <= target_offset" assertion when adapting Content-Range value

[Matus UHLAR - fantomas]

On 02.10.19 16:22, Matus UHLAR - fantomas wrote:
>related to http://bugs.squid-cache.org/show_bug.cgi?id=4977
>
>   (In reply to Matus UHLAR - fantomas from comment #12)
>   > We notice the same problem after few minutes running squid 4.6 (debian 10,
>   > x86_64):
>   >
>   > 2019/10/02 05:00:53 kid1| assertion failed: stmem.cc:98: "lowestOffset () <=
>   > target_offset"
>
>   (In reply to Matus UHLAR - fantomas from comment #14)
>   > I would like to note that we have no SSL support and no header modification
>   > besides "forwarded_for delete"
>
>   This bug report was about adaptation (or, more precisely, new adapted message
>   creation) triggering the above assertion. It sounds like your environment is
>   very different -- no new messages are created. The bug may or may not be the
>   same, of course.
>
>   Can you reproduce the bug with specific message(s), on demand?
>
>I could try to reproduce by using newer squid, hoping the customer won't get
>very angry.  I would need some hints on what more to watch for and what more
>to log. (debug options probably)

OK, now I've got 1.7G pcap,  2.8M access.log (22k requests) and cache.log
saying when did the problem appear.
Any idea what to search for?

-- 
Matus UHLAR - fantomas, uhlar at fantomas.sk ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
Windows found: (R)emove, (E)rase, (D)elete