Build failed in Jenkins: 3.2-matrix » rio.treenet #121 from noc_at_squid-cache.org on 2011-04-11 (squid-dev)

From: <noc_at_squid-cache.org>
Date: Mon, 11 Apr 2011 13:12:03 +0200 (CEST)

See <http://build.squid-cache.org/job/3.2-matrix/./label=rio.treenet/121/changes>

Changes:

[Amos Jeffries] Import SN.png correctly as binary.

[Amos Jeffries] Regression fix: icons/ FHS compliance

It seems I was working out of an outdated copy of the FHS.
Current standards describe /var/www as deprecated.

Reverting the icons location back to where it started.

[Amos Jeffries] Bug 3183: Invalid URL accepted with url host part of only '@'.

3.0 results in an ICAP segfault handling these URLs.

Newer releases do not segfault as easily, but still accept the invalid
URL and there may be other unknown side-effects.

Makes the URL parser present ERR_INVALID_URL for this edge case.

[Amos Jeffries] Portability: allow GnuRegex to use libcompat min()/max()

[Amos Jeffries] Portability: aio.h suffers from GCC-specific hacks on Linux 64-bit

64-bit file support on 32-bit systems uses GCC-specific #define to convert
the functions to 64-bit API but omits the struct aiocb passed as paremeter.

GCC seems not to mind, but non-GCC compilers barf on the invalid types.

[Amos Jeffries] Bug 3192: comm.cc:216: "fd_table[fd].halfClosedReader != NULL"

Polished request reading code to fix CONNECT double-read assertion.

ConnStateData::flags.readMoreRequests, do_next_read variables, and
ClientSocketContext::mayUseConnection() methods were used (or unused!)
incorrectly or inconsistently.

This change removes all do_next_read variables to simplify the state. Instead,
the renamed ConnStateData::flags.readMore indicates whether client_side.cc
should call comm_read. The mayUseConnection() methods are now used to indicate
whether the next client-sent byte (buffered or read) should be reserved for
the current request rather than being interpreted as the beginning of the next
request.

Usually,
flags.readMore mayUseConnection
basic requests: true false
requests with bodies: true true
malformed requests: false false
tunnels: false true

[Amos Jeffries] Bug 3194: selinux may prevent ntlm_smb_lm_auth from using /tmp

[Amos Jeffries] Bug 3185: 3.1.11 fails to compile on OpenBSD 4.8 and 4.9

[Amos Jeffries] Portability Fix: getrlimit() / setrlimit() incompatible type 'struct rlimit'

On Linux (at least) with large file support but not full 64-bit environment.

The getrlimt / setrlimit are #define'd to getrlimite64 / setrlimit64
BUT, the struct rlimit internal fields are updated to 64-bit types individually
instead of a matching #define to struct rlimit64 as a whole.

One can only assume that GCC is casting to void* or some such major voodoo
which hides this type collision.

[Amos Jeffries] Portability: glob.h requires GCC to compile on Linux.

[Amos Jeffries] Enable MemBlob to use the services of MemPools.

Ensure that MemBlobs use MemPools manage memory areas, and remove
duplication of code with MemPools.

[Amos Jeffries] ICC: size_t is guaranteed to be >=0

[Amos Jeffries] ICC: support 64-bit environments dirent definitions

struct dirent is not consistently defined for 32-bit and 64-bit enabled
environments. Provide a dirent_t type defined appropriate to the environment
for use instead.

[Amos Jeffries] Fix ModPoll signedness checks

This npending test bug was preventing any poll() errors from being
noticed and displayed. Possibly leading to some of the weird hanging
reports we have been unable to replicate.

[Amos Jeffries] Extend the number and size of String MemPools

Define and enable 512-bytes, 1kb, 4kb and 16kb mempooled strings
to cover expected upcoming uses

[Amos Jeffries] Fixed chunked request forwarding in ICAP REQMOD presence.

ICAP prohibits forwarding of hop-by-hop headers in HTTP headers. If the virgin
request has a "Transfer-Encoding: chunked" header, the ICAP server will not
receive it. Thus, when the ICAP server responds with a 200 OK and what it
thinks is a copy of the HTTP request, the adapted request will be missing the
Transfer-Encoding header.

One the server side, Squid used to test whether the request had a
Transfer-Encoding header to determine whether request chunking is needed when
talking to the next HTTP hop. That test would fail in ICAP REQMOD presence.

This change implements a more direct/robust check: if we do not know the
request content length, we chunk the request.

We also no longer forward the Content-Length header if we are chunking. It
should not really be there in most cases, but an explicit check is safer and
may also prevent request smuggling attacks via Connection: Content-Length
tricks.

[Amos Jeffries] Portability: Provide stdio wrappers for 64-bit in cstdio C++ builds

stdio.h in that case provides fgetpos64, fopen64 if
__USE_FILE_OFFSET64 is defined. It then checks whether a gcc-specific
__REDIRECT macro is available (defined in sys/cdefs.h, depending on
__GNUC__ begin available).

If it is not available, it does a preprocessor #define.

Which <cstdio> undefines, with this comment:
"// Get rid of those macros defined in <stdio.h> in lieu of real functions.".
When it does a namespace redirection ("namespace std { using ::fgetpos; }")
it goes blam, as fgetpos64 is available, while fgetpos is not.

To fix it, we need to supply global functions matching those
signatures (not macros).

e.g.

#include <stdio.h>
#if defined(__USE_FILE_OFFSET64) &&!defined(__REDIRECT) && defined(fgetpos)
#undef fgetpos
int fgetpos (FILE * f, fpos64_t *p) { return fgetpos64(f,p); }
#endif
#include <cstdio>

This every time we use <cstdio> (directly or indirectly).

This is triggered by us defining -D_FILE_OFFSET_BITS=64 when
--enable-large-files configure is used.

[Amos Jeffries] Enable string mempools to work correctly during initialization phase

Makes string mempools work before Mem::Init() was called, as may happen
during global variable initialization or early main.cc processing. If
needed, strings allocated before the Mem::Init() call are given an extra
buffer space to make sure the allocated buffer size will not match any
string pool size during deallocation.

Shortcomings: We now waste RAM on buffer increase for early allocated
strings unless they are already bigger than the maximum supported string
pool size. Statistics for early allocations are broken. Non-string
mempools still do not support early allocations.

[Amos Jeffries] Fixed %dt logging in the presence of REQMOD.

We use LogEntry::request to save a virgin request for future logging. However,
when that request is adapted and replaced, the adapted request has all the
stats while the saved virgin request lacks them. We have already copied error
details from the adapted to logged/virgin request. Now we copy the DNS wait
time (%dt) as well.

TODO: Move statistics to a stand-alone history object that adapted and
virgin requests can share. Longer term, we should separate HttpRequest
from Master Transaction so that we can store virgin request details without
implicitly storing not-yet-collected master transaction stats.

[Amos Jeffries] Display ERROR in cache.log for invalid configured paths

The validator that checks system paths for files and directories in the
configuration file sends error messages to stderr. It should send them to
cache.log for the admin to see easily.

Also, this makes the error display as FATAL ERROR when using -k parse to
indicate that it is fatal to the startup. Other management signals where
it is not necessarily fatal will only display as an ERROR.

This archive was generated by hypermail 2.2.0 : Mon Apr 18 2011 - 12:00:06 MDT