The procmgmt library replaces the C RunTime Library's signal handlers
but does not restore them on process detachment. This leaves the
process with signal handlers pointing to invalid code that generates
an invalid access error during process termination if the library
was previously unloaded.
potential reclaim in progress fix, and per Chaskiel,
"I don't remember why I put it there, but the fact that
it gets triggered means that we're leaking a vcache object lock. It looks
like the "rename to .__afsXXXX" codepath is responsible (as afsrename does
not use the fact that adp (or aodp) is locked by afs_remove, and locks it
again. I'm surprised it's not deadlocking)" so i coded up a fix
====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================
chaskiel says
The RHS shouldn't be a double negative...
There's no bug (other than the assert itself)
Add new man pages for livesys and voldump. Fix the man page for sys to say
what it actually does, rather than implying that it works like livesys, and
to recommend livesys instead. Fix a path error in the NetInfo
documentation. Update the README for the current status, including
listing all installed commands that don't have man pages. (There may still
be some subcommands that don't have man pages but aren't listed.)
On installation, substitute the configured paths into the man pages,
replacing the Transarc paths. Also fix a problem with the way that
pinstall was being used to install man pages. (Silly me, I was assuming
it had the same behavior as install.)
This is just a quick first pass. Longer term, it's probably better to
replace all paths in the man pages with unambiguous tokens and then
replace those tokens instead of assuming that the man pages use Transarc
paths and replacing those paths specifically. The current method has a
few minor problems, such as not being able to distinguish between the
various paths that make up /usr/afs/bin. Still, the results of this method
are good enough to start with.
Move man page generation out into a separate script that's just invoked
from regen.sh, so that someone can run that separate script later if they
wish. Make that script more robust against problems such as empty podN
directories. Diagnose a missing pod2man and warn about old versions of
Pod::Man.
Also, remove the old programs used to do the initial conversion from HTML.
Enough post-conversion editing was done that they're no longer necessary
except for historical curiosity, and for that purpose they can be pulled
out of CVS.
SIZE_MAX should be UINT_MAX if not defined
====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================
64-bit type safety changes required for successful compilation
on Windows 64-bit systems with the VS 2005 compiler
This patch separates the client and viced capabilties;
adds a viced capabilities for 64bit files;
and modifies the Windows client to probe the viced capabilities
the first time the client discovers the server and uses the capabilities
RPC instead of the GetTime RPC to probe for Up/Down status
====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================
define the capabilities constants
when collecting rx statistics in response to an RPC query, the rx library
attempts to enumerate the number file descriptors in use. This is fine
except that file descriptors are a C Run Time Library concept on Windows
and are not related to networking. In Visual Studio 8, the run time library
will assert() if an invalid file descriptor is passed to fstat() which is
the test used to determine if a file descriptor is valid.
This patch simply returns 0 for the number of file descriptors in use
because that is what would have been returned anyway with the existing
code. What we probably want to return is the number of open socket
handles.
FIXES 24267
include rx.h in order to safely include rx_prototypes.h in order
to obtain a prototype for osi_Alloc() in order to prevent pointer
truncation on systems with 64-bit pointers and 32-bit int.
This patch extends the krb.conf file allowing the specification of
multiple realms which should be treated as equivalents to the local
cell authentication domain. Additional realms are specified on the
first line of the krb.conf file and are separated by white space.
In addition, the patch adds a new file stored in the same directory
as the krb.conf file called krb.excl. This file contains a list of
principal names, one per line, that must not be treated as local
identities.
The purpose of this patch is to allow organizations that are supporting
multiple realms with synchronized user principal databases to allow
their users to login with any of the realms and treat the principal
names as equivalent to the local PTS identity. The exclusion is
to allow certain names, such as those for administrative IDs, to be
restricted to a subset of the realms.
Further optimization of the afs_krb_exclusion() should be performed to
remove the need to re-read the file. This patch should be considered
a temporary solution until a more permanent set of extensions to the
PT database and RPCs allow for the assignment of mechanism specific
aliases for PT IDs.
====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================
correction to exclusion list parsing
If osi_InitDebug() fails due to RPC_S_NO_PROTSEQS, log to the afsd_init.log
file a warning indicating that the registry should be examined for a
misconfiguration of Windows.
This completes the first editing pass of the man pages. Very little
content editing has been done, but the server and client versions of
various man pages have been combined into a single man page for the
file (affects CellServDB, ThisCell, NetInfo, and NetRestrict), the
descriptions of the various AFS cache files have been combined into one
afs_cache man page, and the descriptions of the two butc log files have
been combined into one butc_logs man page.
For man pages for databases with two files, symlinks are now created on
installation for the secondary file name.
All of the man pages should now be ready for public review, additional
editing and cleanup, and content editing.
The attached patch makes the cbd program (the stand-alone version of
the fileserver's callback.c) work again.
(Now I expect 99% "what's that?" and 0% "what a relief!" reactions. As
a reminder that's one of the things you might go hunting for after a
'kill -XCPU' to the fileserver - this program is probably used
infrequently but the older you get... )
This completes the initial editing pass of the section eight man pages.
Only small amounts of content editing has been done. Some known problems
have been noted in README, but there will doubtless be others, as well as
some lingering formatting problems. However, the quality should now be
good enough for general public review.
Some of the section eight man pages were really supposed to be section one,
the package apropros and package help commands are too useless to document,
and a few of the difficult-to-name section five man pages have now acquired
names.
Initial documentation for the man page project, including initial notes
on conversion, a start at a formatting guide, information on how to
contribute, and an initial issues list of things I happened to notice
while editing the section one pages.
Generate the man pages in man1, man5, and man8 subdirectories rather than
directly in the doc/man-pages directory to reduce clutter. Add a
.cvsignore to reduce noise.
Complete an initial editing and cleanup pass for all section one man pages.
Fix various conversion problems, formatting inconsistencies, and obvious
problems. Please note that no editing for content has yet been done; this
is solely editing for formatting and correct conversion to POD.
Also, add some additional section five man pages that were omitted from the
first conversion run due to unusual file names, and globally replace
CAVEATS with CAUTIONS in the man pages to match the original section name.
The section one man pages should now be in reasonable shape and ready for
additional review and further updates, although there are probably still
remaining obvious problems.
====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================
This file got the wrong name when it was originally committed. Fix.
This is the initial conversion of the AFS Adminstrators Reference into POD
for use as man pages. The man pages are now generated via pod2man from
regen.sh so that only those working from CVS have to have pod2man
available. The Makefile only installs. The pages have also been sorted
out into pod1, pod5, and pod8 directories, making conversion to the right
section of man page easier without maintaining a separate list and allowing
for names to be duplicated between pod5 and pod1 or pod8 (which will likely
be needed in a few cases).
This reconversion is done with a new script based on work by Chas Williams.
In some cases, the output is worse than the previous POD pages, but this is
a more comprehensive conversion.
This is only the first step, and this initial conversion has various
problems. In addition, the file man pages that didn't have simple names
have not been converted in this pass and will be added later. Some of the
man pages have syntax problems and all of them have formatting errors. The
next editing pass, coming shortly, will clean up most of the remaining
mess.
this patch replaces the single address in the Interface structure
with an AddrPort structure. This enables the servers to send
messages to the correct port if the port on the Interface does not
match the port stored in the client itself.
add logic to process VNOVNODE in cm_Analyze. Force re-evaluation of symlink
strings and flush the stat cache entry.
force the use of new rx_connections when the server is marked down.
prevent server objects from being freed if user preferences are set.
====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================
improvements to VNOVNODE fix
