2.0a1 : released March 10, 2000
* mod_cgid and suexec have a problem co-existing. suexec sees a null
command string sometimes.
* core dump from 20010418 running 2_0_16
#2 0x8065299 in ap_log_assert (szExp=0x80aaa60 "total_bytes_left > 0 && tmplen > 0", szFile=0x80aa2aa "
core.c", nLine=2555)
#3 0x8075227 in sendfile_it_all (c=0x81470fc, fd=0x814759c, hdtr=0xbfbff670, file_offset=1929216, file_bytes_left=261949,
total_bytes_left=261949, flags=0) at
core.c:2555
#4 0x80761e2 in core_output_filter (f=0x814737c, b=0x814764c) at
core.c:3172
#5 0x806d227 in ap_pass_brigade (next=0x814737c, bb=0x81e80fc) at
util_filter.c:240
#6 0x805e696 in check_pipeline_flush (r=0x820803c) at
http_request.c:388
#7 0x805e707 in ap_process_request (r=0x820803c) at
http_request.c:432
#8 0x805a1a9 in ap_process_http_connection (c=0x81470fc) at
http_core.c:280
#9 0x806bc60 in ap_run_process_connection (c=0x81470fc) at
connection.c:82
#10 0x806be84 in ap_process_connection (c=0x81470fc) at
connection.c:216
#11 0x805fbba in child_main (child_num_arg=272) at
prefork.c:807
#12 0x805fd20 in make_child (s=0x80c64fc, slot=272) at
prefork.c:880
#13 0x805ffec in perform_idle_server_maintenance () at
prefork.c:1021
#14 0x80603d1 in ap_mpm_run (_pconf=0x80c600c, plog=0x80f300c, s=0x80c64fc) at
prefork.c:1191
#15 0x80660cd in main (argc=1, argv=0xbfbffadc) at
main.c:425
#16 0x8059bf9 in _start ()
The input data (received in one read from TCP layer):
Via: 1.0 MDRPRXY01, 1.0 NS2
User-Agent: Mozilla/4.0 (compatible; MSIE 5.5; Windows NT 4.0)
Accept-Language: en-us,tscii;q=0.5
Accept-Encoding: gzip, deflate
The confusion was because apr_sendfile() returned APR_SUCCESS
but zero bytes sent. Presumably the FreeBSD kernel sendfile()
did the same thing (not 100% sure).
Also happened on 20010605...
Accept-Encoding: gzip, deflate
User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)
Again, it would seem that FreeBSD sendfile() returned rc 0 with
no bytes sent. (Other eyes welcome, of course... make sure you
* core dump from 20010521 and 20010529 running 2_0_16 - the "3030" problem
#0 0x80987e8 in apr_cvt (arg=1.3980432860952889e-76,
ndigits=808464432, decpt=0x30303030,
sign=0x30303030, eflag=808464432,
buf=0x30303030 <Address 0x30303030 out of bounds>) at
apr_snprintf.c:177
Cannot access memory at address 0x30303030.
In both coredumps the request is /server-status?auto.
It is unclear whether the apr_*printf function was passed bad
data or it screwed up on its own. 0x30 is '0'. There is a
string of 200-300 '0' characters in the dump, apparently
overlaying enough of the stack to cause serious problems :)
* There is a bug in how we sort some hooks, at least the pre-config
hook. The first time we call the hooks, they are in the correct
order, but the second time, we don't sort them correctly. Currently,
back when this is fixed. rbb
* The AddInputFilter and AddOutputFilter directives do not allow the
administrator to remove or reorder filters. Once a filter is added
in a container, it is present in any subcontainers. It can only be
added to the filter chain after any filters specified in enclosing
* remove the --disable-shared from the subdir config of APR(UTIL)
before the final release. (in fact, it might even be nice to
Note: we need to do a "make install" for APR(UTIL) so the shared
libraries can be installed properly. We could also use that
point to install include files (rather than have Apache
know everything that needs to be installed from the
sub-packages). The original impetus for doing the
disable-shared was because the shared lib wasn't getting
installed and a "make clean" in aprutil would make Apache
* ap_vrprintf() needs to handle more than 4K
* mod_dir should normally redirect ALL directory requests which do
not include a trailing slash on the URI. However, if a "notes"
flag is set (say, via BrowserMatch), this behavior will be
disabled for non-GET requests.
MsgId: <20010227104646.E2297@lyra.org>
MsgId: <3A9C0097.9C83F07C@Golux.Com>
* Usability: Sanitize the MPM config directives. MaxClients in
the threaded MPM is totally misleading now as it has little to
do with limiting the number of clients (it limits the number
of child processes). Bill proposed nomenclature change to
something like "StartWorkers, MaxWorkers, etc." that could
apply to most all the MPMs (with some notable exceptions).
Bill would be happy with changing MaxClients to MaxServers
to make it agree with the operation of the StartServers
RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
* Win32: Rotatelogs sometimes is not terminated when Apache
goes down hard. FirstBill was looking at possibly tracking the
child's-child processes in the parent process.
* Win32: Add a simple hold console open patch (wait for close or
the ESC key, with a nice message) if the server died a bad
death (non-zero exit code) in console mode.
Resolution: bring forward same ugly hacks from 1.3.13-.20
* Port of mod_ssl to Apache 2.0:
step is to figure out how the old three configuration contexts (global,
per-server, per-directory) can be ported to Apache 2.0 (especially the
global context which has to survive server restarts RSE still does not
know how to port). Then the remaining source files (which depend on the
configuration contexts) which are still tagged with "-" in
One more thing to figure out is how and when mod_ssl can provide the
interactive pass phrase dialog (in Apache 1.3 it used the first init
round [where tty was still not detached] and skipped the second init
round [where it was already detached]). Apache 2.0 requires (or already
has?) an official hook where such tty-dialogs can be performed.
RSE is on holiday until June 11th, 2001. After this he starts
working on mod_ssl again and tries to make mod_ssl running inside
Apache 2.0 by end of June. If anyone wants to help making mod_ssl
running in the meantime, feel free to make your hands dirty inside
modules/ssl/. Do whatever you think is appropriate to get it
running. I've no objections of any kind, except: just do not
remove any old functionality. Keep everything, even if it needs
#if
0...endif wrapped to not make trouble for you.
* Performance: Get the SINGLE_LISTEN_UNSERIALIZED_ACCEPT
optimization working in threaded. prefork's new design for how
to notice data on the pod should be sufficient.
* mod_tls is very specific to OpenSSL. Make the API calls
more generic to support other encryption libraries.
* Performance & Debug: Eliminate most (and perhaps all) of the
malloc/free calls in the bucket brigade code. Need some
light weight memory management functions that allow freeing
memory (putting it back into a memory pool) when it is no
longer needed. Enabling simple debugging features like guard
bands, double free detection, etc. would be cool but certainly
Status: Cliff, David, et al have discussed using the blocks SMS
We could then have a thread-private SMS that is pointed
to by the conn_rec's or something so that all calls to
the bucket create functions can pass in that SMS. No locks
required. Should be fast...
* Eliminate unnecessary creation of pipes in mod_cgid
* the autoconf setup should be fixed to default to using the
in a layout should be overridable on the command line. Plus,
what we do right now just doesn't seem to fully fit into how autoconf
works, eg. AC_PREFIX_DEFAULT issues.
Message-ID: <Pine.BSF.4.20.0104031557420.20876-100000@alive.znep.com>
* Combine log_child and piped_log_spawn. Clean up
http_log.c. * Document mod_file_cache.
* OS/2: Make mod_status work for spmt_os2 MPM.
* Platforms that do not support fork (primarily Win32 and AS/400)
Architect start-up code that avoids initializing all the modules
in the parent process on platforms that do not support fork.
would eliminate some code in the Win32 branch that essentially
duplicates what is in APR.
* There are still a number of places in the code where we are
losing error status (
i.e. throwing away the error returned by a
system call and replacing it with a generic error code)
* Mass vhosting version of suEXEC.
the dbmmanage employs the first-matched dbm format. This is not
necessarily the library that Apache was built with. Aught to
rewrite dbmmanage upon installation to bin/ with the proper library
* use apu_dbm in mod_auth_dbm
Status: Greg +1 (low-priority volunteer)
Some additional items remaining:
- case_preserved_filename stuff
(use the new canonical name stuff?)
- find a new home for ap_text(_header)
- is it possible to remove the DAV: namespace stuff from util_xml?
- mod_dav_fs gets built static when mods-shared=most is specified.
* ap_core_translate() and its use by mod_mmap_static and mod_file_cache
are a bit wonky. The function should probably be exposed as a utility
function (such as ap_translate_url2fs() or ap_validate_fs_url() or
something). Another approach would be a new hook phase after
"translate" which would allow the module to munge what the
translation has decided to do.
Status: Greg +1 (volunteers), Ryan +1
* Explore use of a post-config hook for the code in
http_main.c which
calls ap_fixup_virutal_hosts(), ap_fini_vhost_config(), and
ap_sort_hooks() [to reduce the logic in main()]
* read the config tree just once, and process N times (as necessary)
* (possibly) use UUIDs in mod_unique_id
and/or mod_usertrack
* (possibly) port the bug fix for PR 6942 (segv when LoadModule is put
into a VirtualHost container) to 2.0.
* APR-ize resolver stuff in mod_unique_id (Jeff volunteers)
* callers of ap_run_create_request() should check the return value
for failure (Doug volunteers)
* when prefork bails out due to an initialization error in the detached
process (
e.g., mutex init failure), other children (cgid, at least)
* Win32 should grow version info on all .dll/.so/.exe binaries. Pretty
simple with the right script (awk or perl), we just don't need to grow
dozens of new files in the repository to do so.
Suggested by: Jerker Back <info.jelmar@telia.com> PR 7388
* Win32: Get Apache working on Windows 95/98. The following work
(at least) needs to be done:
- winnt MPM: Fix 95/98 code paths in the winnt MPM. There is some NT
specific code that is still not in NT only code paths
- IOL binds to APR sendfile, implemented with TransmitFile, which
is not available on 95/98.
- Document warning that OSR2 is required (for Crypt functions, in
rand.c, at least.) This could be resolved with an SSL library, or
randomization in APR itself.
actually works) and add in a splash of Win9x service code.
PRs that have been suspended forever waiting for someone to
put them into 'the next release':
missing call to "setlocale();"
Additional status for XBitHack directive
Questionable performace of mod_dir() with negotiation
Mod_proxy doesn't allow change of error pages
Modified PATH environemnt variable is not passed, instead
Proxy doesn't deliver documents if not connected
proxy converts ~name to %7Ename when name starts with a dot (.)
mod_access syntax allows hosts that should be restricted
~UserHome directories are not honored in absolute pathname
More LogFormat directives
Proxy FTP Authentication Fails
A smarter "Last Modified" value for SSI documents (see PR number 600)
Request of "Options SymLinksIfGroupMatch"
A security tweak I've been using for a few years for SSI
Proxy doesn't do links right for OpenVMS files through ftp:
imap should read <MAP><AREA>*</MAP> too!
RLimitCPU and RLimitMEM don't apply to all children like they should
Uses cwd before filling it in, doesn't use syslog
it is useful to allow specifiction that root-owned symlinks
should always be followed
Controlling Access to Remote Proxies would be nice...
Adding authentication "on the fly" through the proxy module
request_config field in request_rec is moderately bogus
DoS attacks involving memory consumption
Logging of virtual server to error_log as well
ProxyRemote make a dead cycle.
suexec does not parse arguments to #exec cmd
Allow for Last-Modified: without resorting to XBitHack
insufficent AllowOverrides granularity for autoindexing
improvements to child spawning API
``nph-'' not honored (no buffering) for ProxyRemote mapping
Apache cannot handle continuation line in headers
setlogin() is not called, causing problems with
e.g. identd
regerror() exists, use it
there is no way to keep per-connection per-module state
Add frame-safe anchor attribute to mod_autoindex links
CGI scripts running as Apache user: security (suexec etc.)
add allow,
deny/deny,allow warning to mod_access
Need to know "hit-rate" on proxy cache
Selective url-encode of log fields (or maybe a pseudo
I make mod_headers to modify request headers as well as
No HTTP_X_FORWARDED_FOR set...
ProxyRemote proxy requests fail authentication by firewall
ReadmeName and HeaderName don't allow for server-parsed html.
mod_rewrite forms REQUEST_URI different than mod_cgi does
mod_headers should allow mod_log_config-style formats in
mod_proxy to support persistent conns?
patches to mod_include to allow for file tests
Suggestion for improving authentication modules and core source
code, problem with 401 and ErrorDocument
More Control over autoindex layout
listing of proxy cache content
Allow modules to set user:group for execution.
adding auth_why to conn_rec
pipelined connections are not logged correctly
mod_rewrite doesn't pass Proxy Throughput on internal subrequests
HTTP Server Rebuild Line Needs Changing for the better
mod_status always displays 256 possible connection slots
Make online documentation search link back to my installation
Can not POST to ErrorDocument - Apache/1.3b6
Status module averages are for entire uptime
suexec for general access of user content?
Proposal for TimeZone directive
/server-info doesn't check for the virtual host to list the info
problem specifying ndbm library for build ?with autoconfigure
A small addition to
rotatelogs.c to improve program functionality.
AllowOverride FileInfo is too coarse
TimeOut applies to output of CGI scripts
<IfDenied> directive wanted
CGI's for general use still have to be run as another user
Cache file names in Proxy module
mailto tags and bundling bug report script
Support for System Resource Controller
When will Apache support P3P? Any Plans?
Propose that Apache recommend $UNIQUE_ID for all "session id"
suggestion: power up your Include directive :)
cannot limit some HTTP methods
No module specific data hook for per-connection data
* PR#3191: mod_negotiation
no way to set global quality-of-source (qs) coneg values
* PR#3430: mod_negotiation
Enhancement: MultiViews, Multi-Language Documents
Accessing URL through proxy server corrupts data.
Please add an Apache icon to the win32 systray instead of a DOS window
Some anonymous FTP URLs ask for authentication
BORDER=0 makes Icons look nicer (FancyIndexing)
New ErrorDocumentMatch directive
Need to be able to override shebang line to make CGI scripts
"Files" and "FilesMatch" regexp does not recognize bang as
Please allow CGI env variables (QUERY_STRING, ...) to be logged
Suggestion for better handling of Last-modified headers
mod_cgi prevents handling of OPTIONS requests
[PATCH] install as win32 service with domain account
AllowOverride should have a 'CheckNone' and 'AllowNone' argument
MIME types for MNG and JNG files need adding to
mime.types and
Status: Waiting for IANA types to be defined
Other bugs that need fixing:
* MaxRequestsPerChild measures connections, not requests.
Until someone has a better way, we'll probably just rename it
"MaxConnectionsPerChild".
* Regex containers don't work in an intutive way
Status: No one has come up with an efficient way to fix this
behavior. Dean has suggested getting rid of regex containers
* SIGSEGV on Linux (glibc 2.1.2) isn't caught properly by a
sigwaiting thread. We need to work around this, perhaps unless
there is hope soon for a fixed glibc.
needed. Apache 1.3 just never stashed "multipart" into
r->content_type. We should probably follow suit since the
byterange stuff doesn't want the rest of the code to see the
multipart content-type; the other code should still think it is
dealing with the <orig_ct> stuff.
Status: Greg volunteers to investigate (esp. since he was most
likely the one to break it :-)
Other features that need writing:
* Finish infrastructure in core for async MPMs
* TODO in source -- just do an egrep on "TODO" and see what's there
Documentation that needs writing:
* Mod_status docs are needed.
* The concept of MPMs, especially if we ship more than one MPM for a
* New directives in the various MPMs and appropriate links from
obsolete directives in
core.html to the MPM documentation.
Status: Ben Laurie has written some hooks documentation
* Changes since 1.3.9 can be more easily seen in the commitlog file
which includes some of Roy's comments when the changes were
committed in rough change-sets by purpose. Note that the commitlog
does not show the contents of new files until later.
* Jon Travis's <jtravis@covalent.net> patch to deal with thread-safe
issues with inet_ntoa. See message <20001201163220.A12827@covalent.net>
Status: This is being set aside until the IPv6 work is finished
so that we know exactly what is required.
* Martin Sojka <msojka@gmx.de>'s patch to add error reporting for failed
htpasswd actions due to a full /tmp volume (other programs may have
* Mike Abbott's <mja@trudge.engr.sgi.com> patches to improve
Status: These were written for 1.3, and are awaiting a port to
* Jim Winstead's <jimw@trainedmonkey.com> patch to add CookieDomain and
other small mod_usertrack features
* Dan Rench's <drench@xnet.com> patch to add allow the errmsg and timefmt
of SSI's to be modified in the config file. Patch is available in
* What do we do about mod_proxy?
* Which MPMs will be included with Apache 2.0?