Ben Kaduk [Mon, 15 Oct 2012 17:50:58 +0000 (13:50 -0400)]
Remove stale implement document
As with the texinfo implementor's guide, it is sufficiently stale
so as to be untrustworthy without verification.
Content of this nature should live on k5wiki.kerberos.org.
Ben Kaduk [Mon, 15 Oct 2012 17:38:05 +0000 (13:38 -0400)]
Remove the last vestiges of the texinfo docs
We needed these last stragglers to build the NOTICE file at the
root of the tree, but we now generate a notice.txt from reST.
Man pages have been generated from reST for some time now, we no
longer need scripts to turn them into HTML and ps, as those can
be generated directly from the reST source.
The Makefile only existed to serve the texinfo builds, so it goes
away, too. (The tgz target appears to have been broken since 1996.)
Ben Kaduk [Mon, 15 Oct 2012 21:47:33 +0000 (17:47 -0400)]
Make rules for building NOTICE from RST
Sphinx-build wants to scan over all the sources in the source dir,
and its text output engine does not handle the spanning row cells
used in some of our fancy tables, so it will crash if we attempt
to generate text output from our full suite of documents.
We could work around this by attempting to just compile the one
source file as an additional argument to sphinx-build, but this
seems to insist on producing an index.txt as well as a notice.txt.
Furthermore, we cannot use the same conf.py as in rst_composite because
the paths substitutions are not available with just the minimal
contents of rst_notice. Make a copy and remove the irrelevant parts,
for now, changing only the master_doc field as appropriate.
For the HTML build, we use the standard |copy| macro from isonum.txt
to get the unicode copyright symbol glyph, but for ASCII output we
prefer to just use the literal "(C)".
Ben Kaduk [Fri, 12 Oct 2012 21:12:55 +0000 (17:12 -0400)]
Add notice.rst
To replace the texinfo sources previously used to generate NOTICE.
When compiled to HTML, compares well against the original NOTICE;
use notice.rst in mitK5license.rst instead of using a literal
include of the old NOTICE file.
Ben Kaduk [Fri, 12 Oct 2012 16:42:10 +0000 (12:42 -0400)]
Remove texinfo implementors guide
It is sufficiently old and unloved that nothing in it should be
taken as truth without independent verification.
Content of this nature should live on k5wiki.kerberos.org
in the future.
Ben Kaduk [Fri, 12 Oct 2012 15:36:38 +0000 (11:36 -0400)]
Disconnect the texinfo users guide from the build
Its content has been migrated to or superseded by the krb_users
reST documentation.
The texinfo document attempted to have a general introduction to
Kerberos, but it is not quite suitable for the target audience
of the user's guide and will be rewritten.
A few portions of the texinfo document are simply no longer relevant
and do not need to be migrated. In particular:
The krb5-appl utilities are out of scope for this document, as they
have been moved to a separate repository. Coverage of ksu may be
brought back at a later time, though.
The Kerberos Glossary will be expanded and handled separately.
Greg Hudson [Mon, 15 Oct 2012 18:07:09 +0000 (14:07 -0400)]
Bump libkdb5 API and DAL versions
We made two sets of incompatible changes to the DAL and libkdb5 API
during development for 1.11 (master key list simplification and policy
extensions), so increment the appropriate version numbers.
Greg Hudson [Mon, 15 Oct 2012 17:11:18 +0000 (13:11 -0400)]
Wait for correct message in t_iprop.py
The message "Got incremental updates from the master" precedes
actually replaying the updates on the slave. Instead look for
"Incremental updates:" (the statistics message), which happens just
after the updates are replayed.
Also, we don't need to import time now that we're not sleeping.
When adding {str}, {lenstr}, or {data} to trace output, scan for
bytes which might be non-printable, and add them as hex-escaped
versions of themselves if any are found.
Greg Hudson [Sat, 13 Oct 2012 15:40:51 +0000 (11:40 -0400)]
Eliminate races in t_iprop.py
After we start kpropd, read about the initial full dump before making
changes on the master. Avoid prodding kpropd for this read (by
shifting responsibility for the initial prod to the caller) since
kpropd doesn't sleep before its first request.
When waiting for sync, note whether we got a full propagation and
match that up with our expectations.
Use a long polling interval so kpropd doesn't wake up on its own and
confuse the test script with an extra incremental update.
Greg Hudson [Fri, 12 Oct 2012 20:12:48 +0000 (16:12 -0400)]
Make t_iprop.py faster and more robust
Catch SIGUSR1 in iprop-mode kpropd so that we can use it to interrupt
sleeps and make kpropd do an iprop request immediately.
In k5test.py, add prod_kpropd and read_from_kpropd methods to allow
test scripts to send a SIGUSR1 to kpropd and to read its stdout/stderr
output; also allow the test script to specify additional arguments
when starting kpropd.
In t_iprop.py, start kpropd with -d and, instead of sleeping, read
kpropd output until we see an indication that kpropd is in sync with
the master. To avoid delays, prod kpropd before waiting for sync and
after a completed full prop.
Ben Kaduk [Thu, 11 Oct 2012 16:42:05 +0000 (12:42 -0400)]
Disconnect the texinfo admin guide from the build
Its content has been migrated to or superseded by the reST documentation,
essentially entirely in krb_admins.
A few portions of the texinfo document are simply no longer relevant
and do not need to be migrated. In particular:
Information about reporting bugs lives on k5wiki.kerberos.org.
General Kerberos concepts/introduction will be elsewhere in the tree.
We do not need to document the time zones accepted by kadmin.
We do not need a table of the various error codes and strings in our
formal documentation.
A complete description of the layout of our source tree is not useful
or relevant to most Kerberos administrators.
Greg Hudson [Thu, 11 Oct 2012 17:22:29 +0000 (13:22 -0400)]
Fix cast regexp in C style checker
In check_cast, we want to match cast operators with or without spaces
after the closing paren, and then check for spaces after we match.
Also, per the comment, we want to match potential cast operators
followed by an open paren.
Ben Kaduk [Tue, 9 Oct 2012 20:00:53 +0000 (16:00 -0400)]
Improve the ktadd and ktremove synopses
Mention the options on the synopsis line, and do not imply that
the principal argument(s) for ktadd are optional.
reST line blocks are needed to keep the two forms of ktadd on
separate lines.
Ben Kaduk [Fri, 5 Oct 2012 20:44:40 +0000 (16:44 -0400)]
Fix ordered list style
Sphynx outputs class information that corresponds to its generated
basic.css, which we do not include. This results in all lists,
even nested lists, using arabic numerals.
Import the class properties into kerb.css for now.
Ben Kaduk [Fri, 5 Oct 2012 18:35:45 +0000 (14:35 -0400)]
Fix copy/paste errors in dbadmin
We should include the stashsrvpw content in that section, not
the list content. Likewise, the list_policy content instead
of the destroy_policy content.
Ben Kaduk [Fri, 5 Oct 2012 17:04:16 +0000 (13:04 -0400)]
Update kdb5_util example output
This text has not caught up with changes to the utility itself.
As a side effect, our output text box is narrower and does not have
to scroll on as many browser windows.
Ben Kaduk [Fri, 5 Oct 2012 16:44:52 +0000 (12:44 -0400)]
Wordsmith kdb5_util stash -f
The keyfile worth overriding is the one in kdc.conf. Though using
stash -f would override kdb5_util's -sf argument, there is no reason to
pass both flags to the same invocation.
In any case, the "at startup" language is not really correct.
Ben Kaduk [Thu, 4 Oct 2012 22:04:41 +0000 (18:04 -0400)]
Correct kadm5.acl synopsis
The target principal and restrictions arguments are not orthogonal;
a target principal argument must be given in order for a restriction
list to be supplied.
Ben Kaduk [Thu, 4 Oct 2012 22:00:07 +0000 (18:00 -0400)]
Make the kadm5.acl example sane
It is an eggregious security violation to give all admin principals
admin rights and then give all null instances permission to change
the password of the associated admin instance.
While here, don't assume that admin and root are the only non-null
instances, and correct the formatting of an entry with restrictions.
Greg Hudson [Wed, 10 Oct 2012 17:11:46 +0000 (13:11 -0400)]
Fix slow kprop dejagnu test
Fix kpropd -S -t to actually exit after processing one connection (it
was breaking out of the switch statement, not the while loop). Use
the -t when invoking kpropd from the dejagnu test framework;
previously it was unnecessary because kpropd -S -d exited after one
connection. Clear up some confusion in the kprop.exp comments about
whether kpropd is expected to exit.
Ben Kaduk [Thu, 4 Oct 2012 17:42:13 +0000 (13:42 -0400)]
Document TXT records for realm lookup
Even though they are subject to vulnerabilities via DNS spoofing
and we accordingly don't recommend their use, we do have the code
to use them. Just as we document dns_lookup_realm in krb5.conf(5),
document them here.
Ben Kaduk [Wed, 3 Oct 2012 20:44:28 +0000 (16:44 -0400)]
Remove unused texinfo sources
Now that the install guide make rules are removed, nothing references
build.texinfo or install.texinfo any more (other than the tgz target,
which is updated accordingly).
Ben Kaduk [Wed, 3 Oct 2012 20:16:26 +0000 (16:16 -0400)]
Disconnect texinfo install guide from the build
Its content has been migrated to or superseded by the RST documentation,
split amongst krb_build and various sections of krb_admins.
A few portions of the texinfo document are simply no longer relevant
and do not need to be migrated. In particular:
It's 2012; we don't need to specify that we require a C89 compiler.
It's 2012; it will be easy to get enough disk to build krb5.
The KADM5 tests are part of 'make check' and don't need separate
documentation.
Shared library support is not limited to "a few operating systems".
We do not need to document incompatibilities with ancient/dead OSes.
kadmind4 and v5passwdd are no longer relevant.
Ben Kaduk [Wed, 3 Oct 2012 19:56:46 +0000 (15:56 -0400)]
Add section on updating from single-DES
There are, unfortunately, still some single-DES deployments out
there. Try to help them along by documenting a procedure for
migrating to stronger crypto.
The texinfo install guide had a section on "upgrading", but it was
not really suitable for direct import into a RST document. For one,
it gave a high profile to the on-disk incompatibilities in upgrades
to 1.1 and 1.2. It also was driven at upgrading *to* triple-des (or RC4),
which are something of a dead-end. This new text attempts to be more
general and applicable to today's environment.
Ben Kaduk [Wed, 3 Oct 2012 16:29:20 +0000 (12:29 -0400)]
Add a kdb5_util examples for old KDC upgrades
It's a slightly less-contrived use case of the utility than the
other example, which reads more like a usage statement.
Give a motivating sentence before each example, and note that this
new example is not needed in the general upgrade case.
The need to dump/load for upgrades prior to 1.2 was documented in
the texinfo install guide, but not in any RST sources until now.
Greg Hudson [Tue, 9 Oct 2012 18:27:04 +0000 (14:27 -0400)]
De-conditionalize Camellia code
The Camellia enctypes and cksumtypes have received IANA assignments.
Add #defines using those assignments to krb5.h, remove the CAMELLIA
conditional, and enable testing code as appropriate.
The Camellia draft has not received an RFC number yet, so there is no
Doxygen markup for the enctype and cksumtype #defines. That can be
added once the RFC number is known.
Greg Hudson [Mon, 8 Oct 2012 16:18:35 +0000 (12:18 -0400)]
Remove iprop dejagnu test
Both the Python and dejagnu iprop tests are slow since they use sleeps
to give kpropd time to do its work (although we can fix this with some
work). Since the Python tests cover the same ground as the dejagnu
tests, we don't need both.
Nicolas Williams [Mon, 24 Sep 2012 23:04:50 +0000 (18:04 -0500)]
Remove an old, incorrect comment in kpropd.c
We absolutely do not want a parking brake on the kprop protocol as
described in the comment being removed. Instead the kprop command
should be fixed so it doesn't die on error (assuming it even still does
or ever did, neither of which I've checked).
If a kdb5_util load gets killed between rename()ing the new KDB file
into place and resetting the iprop ulog then the ulog can reflect the
pre-load state, which will almost certainly be incorrect.
This matters because we want to impose a timeout on full resyncs in
kpropd when iprop dictates that a full resync is needed, and the
simplest timeout scheme involves signaling the kdb5_util load process.
But also, we want no such races in general.
The fix is simple: re-initialize the ulog before renaming the new KDB
file into place, then proceed as usual. If the ulog is not properly
updated at the end of the load it will at least always result in
subsequent iprop get updates operations always indicating that a full
resync is required.
Nicolas Williams [Thu, 20 Sep 2012 19:37:45 +0000 (15:37 -0400)]
Make kadmind iprop never return UPDATE_BUSY
Currently kadmind allows slaves to poll for updates as often as they
like, but not within 10s of the last update. This means that iprop will
appear to fail to synchronize the KDC at any site whose master KDC
processes at least one write transaction every 10 seconds consistently.
The original intention must have been to throttle iprop clients (slave
KDCs) that poll too often. But UPDATE_BUSY as implemented is not that,
and implementing a throttle would be difficult (requires keeping state
in a table) and mostly useless (admins can manage their poll timers just
fine without a throttle in kadmind).
Nicolas Williams [Mon, 24 Sep 2012 22:48:48 +0000 (17:48 -0500)]
Remove MAX_ULOGENTRIES
If a master KDC uses only a 64-bit libkadm5srv then there is no reason
to impose any limit on ulog size: the practical maximum will be given by
the filesystem and available storage space.
Even when using a 32-bit libkadm5srv the maximum practical ulog size
will be found easily enough when mmap() fails.
Nicolas Williams [Fri, 17 Aug 2012 23:19:31 +0000 (18:19 -0500)]
Use a single global dump for iprop full syncs
Use a global dump (the default dump file) for full syncs for iprop.
When a slave asks for a fullsync we kprop the existing global dump to it
if that is good enough, else we dump the DB and send the new global
dump.
Before this change kadmind would run kdb5_util dump -i... each time a
slave asked for a full dump. This was done in a sub-process,
thankfully, but it was still a waste of time and storage (e.g., if one
has a huge KDB).
Also, long dump times might cause a slave to give up (the timeout for
this is now configurable). But since iprop dumps bear a serial number
and timestamp and since slaves will resync from that point forward, it
doesn't matter if the dump we send a slave is fresh as long as it is
fresh enough (i.e., that its sno and timestamp are in the ulog).
Also:
- Rename dumps into place instead of unlink, create, write (but we
still keep the dump ok files as lock files and as a method of
signaling to kprop that the dump is complete).
Nicolas Williams [Tue, 25 Sep 2012 02:09:17 +0000 (21:09 -0500)]
Improve kpropd behavior in iprop mode
- Make kpropd in iprop mode fork a child to listen for kprops from the
master. The child writes progress and outcome reports to the parent
for each kprop. This fixes a race between asking for a full resync
and setting up a listener socket for it.
- Add runonce (-t) for kpropd do_standalone() too.
- Add a new iprop parameter: iprop_resync_timeout. kpropd will keep
asking for incremental updates while waiting for a full resync to
finish, and will re-request a full resync if kadmind continues to
indicate that one is needed after this timeout passes since the
previous full resync was requested.
- Allow polling intervals less than 10 seconds.
[ghudson@mit.edu: split out debug output changes; note polling interval
change in commit message]
util/cstyle-file.py checks a file for C style issues and displays
line-by-line output. It is not terribly sophisticated, and can
probably be improved upon (e.g. by doing an emacs batch-reindent of
the file and checking for differences in indentation).
util/cstyle.py produces diffs using git, runs the file checker on each
modified C source file in each diff, and displays the output lines
attribute to the diff.
Luke Howard [Sat, 1 Sep 2012 01:08:27 +0000 (11:08 +1000)]
GENC should always export composite names
RFC 6680 requires that gss_export_name_composite begin the output
token with 04 02. So we must produce a composite token even if the
name has no authdata, and be able to consume a composite token with no
authdata attributes.
Simo Sorce [Tue, 28 Aug 2012 14:47:23 +0000 (16:47 +0200)]
Add SPI calls to import objects by mech oid
An interposer mech needs to be able to handle multiple mechanisms.
When importing a mech token for a name, cred, or context, the
interposer mech needs to know the mech type of the token being
imported. To make this work, add SPI calls which accept a mech type
argument.
Simo Sorce [Tue, 5 Jun 2012 12:09:15 +0000 (08:09 -0400)]
Use interposer mechanisms in mechglue functions
Wherever a GSSAPI mechglue function accepts a mech OID from the
caller, use gssint_select_mech_type() to choose the mechanism to use.
Wherever a mechglue function outputs a mech OID to the caller, use
gssint_get_public_oid() or gssint_make_public_oid_set() to expose the
public mech OID.
Simo Sorce [Mon, 11 Jun 2012 15:50:26 +0000 (11:50 -0400)]
Add primitives for using interposed mechanisms
Add gssint_select_mechanism() to determine what mechanism to use for a
caller-specified OID, gssint_get_public_oid() to determine what
mechanism to expose to the caller, and gssint_make_public_oid_set to
translate an array of mech OIDs into a set of public OIDs. In
gssint_get_mechanism(), match interposed OIDs as well as real ones.
Simo Sorce [Tue, 18 Sep 2012 02:00:44 +0000 (22:00 -0400)]
Add support for loading interposer modules
Extend the syntax of the gss mech config file to allow a module type
delimited by triangle brackets. If the module type is "interposer",
flag the mechanism entry as being an interposer mechanism. A module
marked as an interposer is loaded immediately (so it can interpose a
built-in mechanism) and produces a list of OIDs to interpose.
Interposer mechanisms are not exposed to applications.
Tom Yu [Thu, 27 Sep 2012 21:18:18 +0000 (17:18 -0400)]
Cache TGS-REPs too
Changes in r25660 inadvertently failed to insert TGS-REPs into the
lookaside cache. Call finish_dispatch_cache() at the end of
dispatch() to handle this case.
Ben Kaduk [Fri, 21 Sep 2012 17:01:06 +0000 (13:01 -0400)]
Remove kerbsrc.win
It has been unloved and broken repeatedly for many years, requiring
updating of several variables whenever new directories are added
and similar tedia. It was originally intended to avoid the need
for Unix utilities on Windows, but Microsoft provides the Utilities
and SDK for UNIX-based Applications which is enough rope to do a
native build.
Leave behind a warning message to anyone who does try to build the
target.
Clean up some now-unused infrastructure in the build system.
Ben Kaduk [Fri, 21 Sep 2012 15:13:32 +0000 (11:13 -0400)]
Remove kerbsrc-nt
NT is long-gone; we don't need to keep a special-case error message
around telling people not to use it.
Clean out the unneeded code from the Makefile.in
Ben Kaduk [Thu, 20 Sep 2012 22:03:54 +0000 (18:03 -0400)]
Remove kerbsrc83
It's been a long time since systems were limited to 8.3 format
for file names. No one should be thinking to try and build this
target, and if they do, we don't need a custom error message anymore.
Clean out the unneeded code from the Makefile.in.
Ben Kaduk [Thu, 20 Sep 2012 18:03:20 +0000 (14:03 -0400)]
Do not add empty dir to the include search list
windows/leashdll/include contains only a krb4 directory.
We have a dubious need for the latter through AFSroutines.c,
but the former can be eliminated.
Ben Kaduk [Wed, 19 Sep 2012 16:13:30 +0000 (12:13 -0400)]
Update windows/README
The build instructions have changed somewhat, as have the requirements
for a build environment.
The default behavior for KRB5_CONFIG and KRB5CCNAME has also changed.
Attempt to remove mention of overly specific Windows versions that
are now quite old when the behavior persists in newer versions of Windows.
Document the usage of DNS by default and the reduced need for a large
krb5.ini file.
Talk a little more about the LSA cache.
Unlike most GSS test programs, t_s4u2proxy_krb5 uses a cleanup
handler, so we have to be careful to initialize everything we clean
up--particularly service2_name, which is initialized after a possible
goto. Also, remember to release acceptor_name.