summaryrefslogtreecommitdiff
path: root/source3/lib
AgeCommit message (Collapse)AuthorFilesLines
2010-10-04s3: Attempt to get the Tru64 build a bit furtherVolker Lendecke1-1/+0
2010-10-02s3: Attempt to fix bug 7665Volker Lendecke11-136/+295
Quite a few of our internal routines put stuff on talloc_tos() these days. In top-level netapi routines, properly allocate a stackframe and clean it again. Also, don't leak memory in the rpccli_ callers onto the libnetapi context.
2010-10-01s3:events: Call all ready fd event handlers on each iteration of the main loopSteven Danneman1-7/+15
Previously, only one fd handler was being called per main message loop in all smbd child processes. In the case where multiple fds are available for reading the fd corresponding to the event closest to the beginning of the event list would be run. Obviously this is arbitrary and could cause unfairness. Usually, the first event fd is the network socket, meaning heavy load of client requests can starve out other fd events such as oplock or notify upcalls from the kernel. In this patch, I have changed the behavior of run_events() to unset any fd that it has already called a handler function, as well as decrement the number of fds that were returned from select(). This allows the caller of run_events() to iterate it, until all available fds have been handled. I then changed the main loop in smbd child processes to iterate run_events(). This way, all available fds are handled on each wake of select, while still checking for timed or signalled events between each handler function call. I also added an explicit check for EINTR from select(), which previously was masked by the fact that run_events() would handle any signal event before the return code was checked. This required a signature change to run_events() but all other callers should have no change in their behavior. I also fixed a bug in run_events() where it could be called with a selrtn value of -1, doing unecessary looping through the fd_event list when no fds were available. Also, remove the temporary echo handler hack, as all fds should be treated fairly now.
2010-10-01samba: share readline wrappers among all buildsystems.Günther Deschner1-174/+0
Guenther
2010-10-01s3-readline: move cmd_history to smbclient, the only user.Günther Deschner1-21/+0
Guenther
2010-10-01samba: share select wrappers.Günther Deschner6-207/+5
Guenther
2010-09-28s3: Remove talloc_autofree_context() from myhostname()Volker Lendecke1-3/+1
No destructor needed, so we can as well use the NULL context
2010-09-27Change to using TDB_INCOMPATIBLE_HASH (the jenkins hash) on allJeremy Allison6-11/+11
TDB_CLEAR_IF_FIRST tdb's. For tdb's like gencache where we open without CLEAR_IF_FIRST and then with CLEAR_IF_FIRST if corrupt this is still safe to use as if opening an existing tdb the new hash will be ignored - it's only used on creating a new tdb not opening an old one. Jeremy.
2010-09-26s3: Remove talloc_autofree_context() from nametouid()Volker Lendecke1-1/+1
pass is freed a few lines down
2010-09-26s3: Remove talloc_autofree_context() from getpwnam_alloc()Volker Lendecke1-1/+1
This is given to the memcache a few lines down
2010-09-26s3: Remove talloc_autofree_context() from serverid_db()Volker Lendecke1-2/+2
If we needed an explicit tdb_close() in the destructor, we'd be hosed long ago.
2010-09-26s3: Remove talloc_autofree_context() from serverid_parent_init()Volker Lendecke1-3/+2
2010-09-26s3: Remove talloc_autofree_context() from messaging_tdb_parent_init()Volker Lendecke1-3/+2
2010-09-26s3: Remove talloc_autofree_context() from ctdb_read_req()Volker Lendecke1-1/+2
2010-09-23s3-dsgetdcname: always pass in messaging context.Günther Deschner3-4/+22
Volker, please check. Guenther
2010-09-22s3-registry: fix some c++ build warnings.Günther Deschner1-2/+2
Guenther
2010-09-22s3:registry: remove unneeded TALLOC_CTX argument from reg_deletekey_recursiveMichael Adam1-3/+2
2010-09-22s3-lib: add srprs, primitives to build simple recursive parsersGregor Beck2-0/+366
Signed-off-by: Michael Adam <obnox@samba.org>
2010-09-22s3-lib: add cbuf, a talloced character bufferGregor Beck2-0/+522
Signed-off-by: Michael Adam <obnox@samba.org>
2010-09-21s3: Remove a global variable in bugfix for bug 7665Volker Lendecke2-7/+11
Signed-off-by: Günther Deschner <gd@samba.org>
2010-09-21s3:registry: move prototype from reg_init_smbconf.c to its own header.Michael Adam1-0/+1
2010-09-21s3:registry: move the reg_api prototypes to their own header.Michael Adam1-0/+1
2010-09-20s3-libnetapi: Fix Bug #7665, memory leak in netapi connection manager.Günther Deschner1-30/+65
Guenther
2010-09-20s3-lib/util Remove unused smb_uuid_unpack()Andrew Bartlett1-31/+0
Andrew Bartlett Signed-off-by: Günther Deschner <gd@samba.org>
2010-09-20s3-uuid Remove unused smb_uuid_pack()Andrew Bartlett1-9/+0
Signed-off-by: Günther Deschner <gd@samba.org>
2010-09-20s3: Replace sid_binstring and sid_guidstring with PIDL-based alternativesAndrew Bartlett2-33/+0
This reduces the manual marshalling of these structures by removing the duplication here. Andrew Bartlett Signed-off-by: Günther Deschner <gd@samba.org>
2010-09-20s3-util: use shared dom_sid_dup.Günther Deschner4-30/+13
Guenther
2010-09-20s3-util_sid: use shared dom_sid_compare_auth and dom_sid_equal_X functions.Günther Deschner4-90/+15
Guenther
2010-09-20s3-build: only include ctdbd_conn.h where needed.Günther Deschner3-0/+3
Guenther
2010-09-20s3-build: only include async headers where needed.Günther Deschner1-0/+1
Guenther
2010-09-20Really enable core dumps in LinuxRusty Russell1-8/+8
commit e1f1ce68e1f685400a8c68bcec14018e3d1fc29d Author: Rusty Russell <rusty@rustcorp.com.au> Date: Mon Sep 20 13:33:30 2010 +0930 source3: dump core on Linux, even after seteuid/etc. The "dumpable" flag is reset on every call to set*uid, so we need to reset it to 1 in the signal handler itself. This code dates back to commit ac01fda2b97b 6 years ago :( Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Jeremy Allison <jra@samba.org>
2010-09-20s3:registry: move higher level function reg_open_path to new module reg_api_utilMichael Adam1-0/+1
The reg_api.c code should just export functions that direclty relate to winreg api calls.
2010-09-19s3-nltest: add dsregdns command to nltest.Günther Deschner1-1/+21
Guenther
2010-09-19s3-libnetapi: support NETLOGON_CONTROL_FORCE_DNS_REG in I_NetLogonControl().Günther Deschner1-0/+4
Guenther
2010-09-16s3/s4: merge msleep and smb_msleepBjörn Jacke1-49/+0
the merged variant is renamed to smb_msleep as some platforms already have a msleep function.
2010-09-15s3-nltest: add dsgetdc command.Günther Deschner1-3/+181
Guenther
2010-09-15s3-libnetapi: try using netr_DsRGetDCNameEx() in DsGetDcName().Günther Deschner1-0/+13
Guenther
2010-09-15s3-libnetapi: add DS request flags to public header.Günther Deschner1-0/+21
Guenther
2010-09-15s3-libnetapi: add DS_X_FLAGs to public header.Günther Deschner1-0/+16
Guenther
2010-09-15s3-nltest: convert server input argument into --server.Günther Deschner1-24/+19
Guenther
2010-09-15s3-nltest: rename print_result to print_netlogon_info_result.Günther Deschner1-4/+17
Guenther
2010-09-14Ensure incoming timespec values correctly wrap at nsecs.Jeremy Allison1-0/+4
Jeremy.
2010-09-14s3-util_sid Use the NDR parser to parse struct dom_sidAndrew Bartlett1-14/+8
The manual parser failed to constrain the maximum number of sub-authorities to 15, allowing an overflow of the array. Andrew Bartlett
2010-09-14libcli/security Merge source3/ string_to_sid() to common codeAndrew Bartlett1-106/+0
The source3 code repsects the limit of a maximum of 15 subauths, while the source4 code does not, creating a security issue as we parse string-form SIDs from clients. Andrew Bartlett
2010-09-14s3-util_sid use ARRAY_SIZE() to ensure we never overflow the dom_sidAndrew Bartlett1-1/+1
This ensures that this, unlike the MAXSUBAUTHS macro, can't get out of sync with the structure. Andrew Bartlett
2010-09-14s3-util_sid Accept S-1-5 as a SIDAndrew Bartlett1-2/+10
2010-09-14s3-dom_sid Use C99 types in dom_sid handlingAndrew Bartlett1-7/+7
Andrew Bartlett
2010-09-14s3/profiling: don't use CLOCK_PROCESS_CPUTIME_IDBjörn Jacke1-17/+0
that clock is a CPU burnometer but we need a chronometer for profiling.
2010-09-11libcli/security Move 'private' privileges functions to another headerAndrew Bartlett1-0/+1
These functions work on the bitmap, and are only exposed because the source3/ privileges storage uses the bitmap in account_policy.tdb Andrew Bartlett Signed-off-by: Andrew Tridgell <tridge@samba.org>
2010-09-11libcli/security Remove 'always true' return from se_priv_put_all_privilegesAndrew Bartlett1-3/+1
Andrew Bartlett Signed-off-by: Andrew Tridgell <tridge@samba.org>