I have a strange issue with SVN 1.4.4 on Sol 10 and Apache 2.0.58. I
have tried (recompiling against) neon versions 2.25.5 and 2.26.4 with
the same results.
I have everything set up fine and I can browse/commit/checkout over
http/https from a Windows box with TortoiseSVN with no problems so it's
not the server part of SVN that's the trouble.
However, on the SVN server, if I try to use the svn client CLI to
checkout/import or commit over http/https, I get a segfault every time.
"svn list" over http/https seems to work fine. file:// URLs work fine
for everything. I can reliably reproduce this against any http/https URL
so it's certainly a client issue. I've tried re-compiling svn against
two different neon versions (2.25.5 and 2.26.4), static and dynamic - no
difference. A quick dbx debug gives:
(dbx) run co https://some.server/some/path /tmp/blah
Running: svn co https://some.server/some/path /tmp/blah
(process id 1124)
Reading 646%UTF-8.so
Reading UTF-8%646.so
[EMAIL PROTECTED] ([EMAIL PROTECTED]) signal SEGV (no mapping at the fault
address) in (unknown) at 0x0
0x00000000: <bad address 0x0>
Current function is send_request_body
393 if (req->body_cb(req->body_ud, NULL, 0) != 0) {
(dbx) exit
That line is in in neon's ne_request.c
Any ideas?
PK
--
Dr Philip L Kime
_______________________________________________
neon mailing list
[email protected]
http://mailman.webdav.org/mailman/listinfo/neon
