summaryrefslogtreecommitdiff
path: root/source3/smbd/server.c
AgeCommit message (Collapse)AuthorFilesLines
1999-12-13first pass at updating head branch to be to be the same as the SAMBA_2_0 branchAndrew Tridgell1-176/+148
(This used to be commit 453a822a76780063dff23526c35408866d0c0154)
1999-12-12final part of "first" phase converting over to msrpc daemon architecture.Luke Leighton1-11/+0
done a minimal amout of clean-up in the Makefile, removing unnecessary modules from the link stage. this is not complete, yet, and will involve some changes, for example to smbd, to remove dependencies on the password database API that shouldn't be there. for example, smbd should not ever call getsmbpwXXX() it should call the Samr or Lsa API. this first implementation has minor problems with not reinstantiating the same services as the caller. the "homes" service is a good example. (This used to be commit caa50525220b0d0250fa139367593c2de2c12135)
1999-12-08ABOUT TIME!!!!!!!!Luke Leighton1-2/+0
damn, this one is bad. started, at least two days ago, to add an authentication mechanism to the smbd<->msrpc redirector/relay, such that sufficient unix / nt information could be transferred across the unix socket to do a become_user() on the other side of the socket. it is necessary that the msrpc daemon inherit the same unix and nt credentials as the smbd process from which it was spawned, until such time as the msrpc daemon receives an authentication request of its own, whereupon the msrpc daemon is responsible for authenticating the new credentials and doing yet another become_user() etc sequence. (This used to be commit 30c7fdd6ef10ecd35594311c1b250b95ff895489)
1999-12-06the first independent msrpc daemon - lsarpcd.Luke Leighton1-0/+12
one horrible cut / paste job from smbd, plus a code split of shared components between the two. the job is not _yet_ complete, as i need to be able to do a become_user() call for security reasons. i picked lsarpcd first because you don't _need_ security on it (microsoft botched so badly on this one, it's not real. at least they fixed this in nt5 with restrictanonymous=0x2). fixing this involves sending the current smb and unix credentials down the unix pipe so that the daemon it eventually goes to can pick them up at the other end. i can't believe this all worked!!! (This used to be commit 2245b0c6d13c7c5886e81f9137b05df883598c26)
1999-12-01sys_select added one more argument (read, write selectors).Luke Leighton1-1/+1
(This used to be commit e4d92ff9dfc51735e6932748f66a7c20b2c1cb6a)
1999-11-24WARNING! MOVED rpc_server/srv_lsa_hnd.c TO lib/util_hnd.cLuke Leighton1-1/+4
CVS UPDATE MAY ISSUE WARNING ABOUT lib/util_hnd.c MODIFICATION DATE BEING IN THE FUTURE. CVS CHECKOUT A NEW REPOSITORY MAY BE SAFER. (This used to be commit c26e8a6ac59934ae580077df937439c2c538dbdb)
1999-11-24rewrote policy handle code to be generic (it's needed for client-side too)Luke Leighton1-1/+1
attempted to fix regsetsec command (This used to be commit eaac0923e0e5e3f4c3d944272a71f3235ac2a741)
1999-11-18set "key does not exist" response to reg_query_val() (win32 status codeLuke Leighton1-0/+6
of 0x2). [p.s. getting REALLY bored of this nt5rc2->samba domain stuff]. (This used to be commit a8801942f72b2394a54e4a46546029fac70efeed)
1999-10-05David van Geyn spotted that ulong is not defined on FREEBSD, not supposedLuke Leighton1-1/+1
to use ulong use uint32 instead. (This used to be commit 16dd94db1fd2ed8ed7d240c5aac5a481b64cd26d)
1999-09-15#defines for port 445 to SMB_PORT2Luke Leighton1-4/+4
(This used to be commit a8d4560e0064a67a234eae89a564b79d2426d9a9)
1999-09-10oops, port 455 instead of 445Luke Leighton1-1/+1
(This used to be commit 4e5eac18d0d9325c8e524fb53b4e049f0893fae7)
1999-09-08bertl <bs@niggard.org> patch for making samba listen on port 445, just likeLuke Leighton1-28/+48
nt 5 does. cool! (This used to be commit 727e1421b2c0f8c72c428776593498103304082b)
1999-07-12Jean-Francois Micouleau's rewritten DFS patch, originally written byLuke Leighton1-0/+3
Nigel Williams. despite the data format being *exactly* the same as NT's, this still doesn't work yet. more work needed. (This used to be commit 270981960bb5aab52d2f8e494827101ece6729c4)
1999-04-27rpc_parse/parse_misc.c : defined a new BUFFER5 structJean-François Micouleau1-0/+1
include/ntdomain.h : added rpc_spoolss.h include statement include/proto.h include/rpc_dce.h : added definition of RPC_ALTER_CONTEXT request & reply param/loadparm.c : 2 new options for NT printing support and some changes to initial values in the LPRNG case. rpc_parse/parse_prs.c : added prs_uint16s() rpc_parse/parse_rpc.c : added SYNT_SPOOLSS_V1 and code for the alter-context support. rpc_server/srv_pipe.c : alter-context support smbd/nttrans.c smbd/server.c include/rpc_misc.h Makefile.in include/smb.h Jean Francois (This used to be commit 4c515804b70254248e378a3f90f47e4c32639d29)
1999-04-01Ken McDonell from SGI was interested in adding some profilingAndrew Tridgell1-1/+8
capabilities to Samba so that Samba could talk to the SGI PCP (Performance Co-Pilot) apps. This change adds a profiling shared memory area and uses it to count two fairly trivial things, the number of uid switches and the number of SMB packets processes. To add more just edit include/profile.h and then increment it at the right place. I've also added a -P switch to smbstatus to dump the profile area. (This used to be commit ef3d61a80ad0b87c196a63ef2bd4fe8fc3863bd0)
1999-02-08reload_services needs to be called prior to init_files but afterLuke Leighton1-4/+4
get_myname. (This used to be commit 9ab81caa065a0e08368fc9137d42ed810fd4b817)
1999-02-03initialise my name (used in %h) prior to loading smb.conf files.Luke Leighton1-26/+15
(This used to be commit ed128c38a88746ec7822d598e72f0106a30a4af7)
1998-12-01rpc_samr.h parse_samr.c srv_samr.c :Luke Leighton1-1/+1
samr_query_aliasmembers (cool!) util_pwdb.c sids.c nmbd.c server.c smbpasswd.c swat.c : pwdb_initialise(BOOL is_server) now creates / reads DOMAIN_NAME.SID if is_server is True, and does LsaQueryInfoPolicy(levels 3 and 5) to obtain member and pdc sids. (This used to be commit 3e1eb4f26b67e484b05e1dde94fd4e4dae982631)
1998-11-30- adding builtin[alias]db.Luke Leighton1-0/+5
- lib/sids.c: generate_sam_sid() modified to take a domain name: it now generates "DOMAIN_NAME.SID". reasons: 1) if you run multiple samba servers on the same machine under different netbios names as members of a domain, they won't all use the same SID, which is a _big_ mistake but it would happen _by default_. 2) we have (had) a problem with sid_to_string() and string_to_sid() which cause SIDs to be incorrectly read. one of the major reasons for *NOT* making this change was so as not to disrupt existing users. but as they will be anyway by this bug, we might as well go ahead. - passdb/smbpass.c: wanted to change the meaning of the name in the smbpasswd file to an "nt" name not a "unix" name. this is probably not a good idea: reverted this. - output formatting / bug-fixing in rpcclient query_useraliases code. (This used to be commit e4930f5f48f8246ceec8add8bf769954a963190c)
1998-11-29weekend work. user / group database API.Luke Leighton1-18/+19
- split sam_passwd and smb_passwd into separate higher-order function tables - renamed struct smb_passwd's "smb_user" to "unix_user". added "nt_user" plus user_rid, and added a "wrap" function in both sam_passwd and smb_passwd password databases to fill in the blank entries that are not obtained from whatever password database API instance is being used. NOTE: whenever a struct smb_passwd or struct sam_passwd is used, it MUST be initialised with pwdb_sam_init() or pwd_smb_init(), see chgpasswd.c for the only example outside of the password database APIs i could find. - added query_useraliases code to rpcclient. - dealt with some nasty interdependencies involving non-smbd programs and the password database API. this is still not satisfactorily resolved completelely, but it's the best i can do for now. - #ifdef'd out some password database options so that people don't mistakenly set them unless they recompile to _use_ those options. lots of debugging done, it's still not finished. the unix/NT uid/gid and user-rid/group-rid issues are better, but not perfect. the "BUILTIN" domain is still missing: users cannot be added to "BUILTIN" groups yet, as we only have an "alias" db API and a "group" db API but not "builtin-alias" db API... (This used to be commit 5d5d7e4de7d1514ab87b07ede629de8aa00519a1)
1998-11-25Makefile.in: Added maintainer mode fixes.Jeremy Allison1-1/+1
aclocal.m4: Added AC_LIBTESTFUNC. configure.in: Fixed -lsecurity -lsec problems. client.c: dos_ fixes. groupdb/aliasunix.c: Dead code removal. include/includes.h: Added default PRINTCAP_NAME. lib/genrand.c: dos_ fixes. lib/replace.c: Added strtoul. lib/system.c: dos_ fixes. lib/util.c: dos_ fixes. lib/util_sid.c: Signed/unsigned fixes. lib/util_str.c: removed bad const. locking/locking_slow.c: dos_ fixes. printing/printing.c: dos_ fixes. rpc_server/srv_samr.c: Dead code removal. rpc_server/srv_sid.c: global_myworkgroup defined with wrong size AGAIN ! smbd/dir.c: dos_ fixes. smbd/open.c: dos_ fixes. smbd/oplock.c: dos_ fixes. smbd/reply.c smbd/server.c smbd/service.c smbd/uid.c: dos_ fixes. Jeremy. (This used to be commit 6acb4b68f68d516e2ac3c47e500f5600d653435e)
1998-11-25fixing domain join and domain login problemsLuke Leighton1-1/+1
(This used to be commit 90a24664318da97a6e8cfe4622a8573c0e3cbe5e)
1998-11-23unix instance of group database APILuke Leighton1-19/+1
(This used to be commit e76f593b3572ac881f1aa1fb3326d8b7169b0078)
1998-11-17- group database API. oops and oh dear, the threat has been carried out:Luke Leighton1-2/+49
the pre-alpha "domain group" etc parameters have disappeared. - interactive debug detection - re-added mem_man (andrew's memory management, detects memory corruption) - american spellings of "initialise" replaced with english spelling of "initialise". - started on "lookup_name()" and "lookup_sid()" functions. proper ones. - moved lots of functions around. created some modules of commonly used code. e.g the password file locking code, which is used in groupfile.c and aliasfile.c and smbpass.c - moved RID_TYPE_MASK up another bit. this is really unfortunate, but there is no other "fast" way to identify users from groups from aliases. i do not believe that this code saves us anything (the multipliers) and puts us at a disadvantage (reduces the useable rid space). the designers of NT aren't silly: if they can get away with a user- interface-speed LsaLookupNames / LsaLookupSids, then so can we. i spoke with isaac at the cifs conference, the only time for example that they do a security context check is on file create. certainly not on individual file reads / writes, which would drastically hit their performance and ours, too. - renamed myworkgroup to global_sam_name, amongst other things, when used in the rpc code. there is also a global_member_name, as we are always responsible for a SAM database, the scope of which is limited by the role of the machine (e.g if a member of a workgroup, your SAM is for _local_ logins only, and its name is the name of your server. you even still have a SID. see LsaQueryInfoPolicy, levels 3 and 5). - updated functionality of groupname.c to be able to cope with names like DOMAIN\group and SERVER\alias. used this code to be able to do aliases as well as groups. this code may actually be better off being used in username mapping, too. - created a connect to serverlist function in clientgen.c and used it in password.c - initialisation in server.c depends on the role of the server. well, it does now. - rpctorture. smbtorture. EXERCISE EXTREME CAUTION. (This used to be commit 0d21e1e6090b933f396c764af535ca3388a562db)
1998-10-28Quick fixes to fix the broken tree. Needed for my morning compiles.Jeremy Allison1-0/+1
Chris - feel free to fix these things differently if these fixes don't work for you. Jeremy. (This used to be commit 1f225abfbe6b021b589711c2e4c599a63050c5b2)
1998-10-22libsmb/smbdes.c: #ifdef'ed out code prior to removal.Jeremy Allison1-1/+1
rpc_client/cli_pipe.c: Inlined code removed from smbdes.c rpc_server/srv_samr.c: Fixed unused variable warning. rpc_server/srv_util.c: Inlined code removed from smbdes.c Luke - the above changes are the first part of the changes you and I discussed as being neccessary at the CIFS conference. *PLEASE REVIEW THESE CHANGES* - make sure I haven't broken any of the authenticated DCE/RPC code. smbd/nttrans.c: Fixed to allow NT5.0beta2 to use Samba shares with NT SMB support. smbd/open.c: Fixed mkdir when called from nttrans calls. smbd/server.c: Set correct size for strcpy of global_myworkgroup. Jeremy. (This used to be commit d891421d16ff80998dee429227bd391455f9d1a1)
1998-10-21domain aliases added a bit better: does local aliases if you queryLuke Leighton1-2/+2
for sid S-1-5-20 and does (nothing at the moment) if you query for your own sid. (This used to be commit da40f26f4b2f7ce286076b4e39dffd76aa2ef8e6)
1998-09-30libsmb/clientgen.c: Fixed signed/unsigned compile warnings spotted by Herb.Jeremy Allison1-2/+2
param/loadparm.c: smbd/oplock.c: Allow kernel oplocks to be turned off in the smb.conf file. smbd/server.c: Move init_structs() to after the smb.conf file is loaded - preparation for making a "max open files" parameter. Jeremy. (This used to be commit 6a261517a09b005f502a37941431308fa8bf2c5c)
1998-09-25Makefile.in: Fixed bug with continuation line causing proto to fail.Jeremy Allison1-2/+10
Added $(PROGS) $(SPROGS) as targets for make clean. acconfig.h: Added HAVE_IRIX_SPECIFIC_CAPABILITIES. configure.in: Added sys/capability.h header check. Added function checks for srandom random srand rand. Added HAVE_IRIX_SPECIFIC_CAPABILITIES test. includes.h: Added #include <sys/capability.h>. ntdomain.h: Moved struct acct_info into here from smb.h smb.h: Added KERNEL_OPLOCK_CAPABILITY define. Moved enum action_type into rpcclient.h Moved struct cli_state into client.h Moved struct nt_client_info, struct tar_client_info, struct client_info into rpcclient.h lib/genrand.c: Changed to use sys_random() & friends. lib/smbrun.c: Lose capabilities after fork. lib/system.c: Added set_process_capability(), set_inherited_process_capability() sys_random(), sys_srandom(). lib/util.c: Added Ander's EFBIG lock check to fcntl_lock for 64 bit access to an 32 bit mounted NFS filesystem. nmbd/nmbd.c: Changed to use sys_random() & friends. nmbd/nmbd_browsesync.c: Changed to use sys_random() & friends. passdb/ldap.c: Missed one pdb_encode_acct_ctrl call. passdb/passdb.c: Changed to Ander's code for ' ' characters. passdb/smbpass.c: Added Ander's code to reset ACB_PWNOTREQ. script/mkproto.awk: Added 'long' to prototypes. smbd/chgpasswd.c: Lose capabilities after fork. smbd/open.c: Do the mmap *after* the kernel oplock. smbd/oplock.c: Removed stub code from kernel oplock path. Added set_process_capability(), set_inherited_process_capability() calls. smbd/reply.c: Initialize count = 0, offset = 0. smbd/server.c: Added set_process_capability(), set_inherited_process_capability() calls. tests/summary.c: Ensure we have RANDOM or RAND. utils/smbpasswd.c: Added Ander's code to reset ACB_PWNOTREQ. utils/torture.c: Changed to use sys_random() & friends. Jeremy. (This used to be commit e8be306f23963ac00b1a383ebe0cc1421529fb02)
1998-09-23smbd/oplock.c: Use O_CREAT and O_TRUNC and correct mode flagsJeremy Allison1-2/+2
when creating oplock test file. smbd/server.c: Check for existance of kernel oplocks before becoming a daemon. Jeremy. (This used to be commit b42779e17e754d4a2f75904e2187c9209e0a53f0)
1998-09-23First cut at kernel oplocks. This should have no effect unless runninJeremy Allison1-0/+2
on a machine that supports them in autoconf. Move various functions out of lib/util.c into smbd/process.c and smbd/oplock.c where they belong. Jeremy. (This used to be commit c3c5e13f85c97939746070132dad941e79c546fb)
1998-09-09Copyright notices written to debug logs are now through 1998.Christopher R. Hertel1-1/+1
Chris -)----- (This used to be commit ac13c29d46f564fc340b652b4b71dfa92e4b5b16)
1998-09-05tridge the destroyer returns!Andrew Tridgell1-2/+2
prompted by the interpret_security() dead code that Jean-Francois pointed out I added a make target "finddead" that finds potentially dead (ie. unused) code. It spat out 304 function names ... I went through these are deleted many of them, making others static (finddead also reports functions that are used only in the local file). in doing this I have almost certainly deleted some useful code. I may have even prevented compilation with some compile options. I apologise. I decided it was better to get rid of this code now and add back the one or two functions that are needed than to keep all this baggage. So, if I have done a bit too much "destroying" then let me know. Keep the swearing to a minimum :) One bit I didn't do is the ubibt code. Chris, can you look at that? Heaps of unused functions there. Can they be made static? (This used to be commit 2204475c87f3024ea8fd1fbd7385b2def617a46f)
1998-09-02we are never interested in SIGPIPE so just ignore (block) itAndrew Tridgell1-23/+3
always. Don't even install a handler. (This used to be commit 72c383f4b7bcce5374632dc972df16ab0e2542b9)
1998-08-28This checking fixes the statcache bug that stopped NetBench from runningJeremy Allison1-1/+1
correctly. Added new parameter "stat cache size" - set to 50 by default. I now declare the statcache code officially "open" for business :-). It gets a hit rate of 97% with a NetBench run and seems to make using a case insensitive run as efficient as a case sensitive run. Also tidied up our sys_select usage - added a maxfd parameter and also added an implementation of select in terms of poll(), for systems where poll() is much faster. This is disabled by default. Jeremy. (This used to be commit 779b924ec1f6c81ff578d22295b20fece698d1fc)
1998-08-27This is the stat cache code - seems to work fine (needs heavyJeremy Allison1-1/+1
NetBench testing though.... :-). Attempts to efficiently reduce the number of stat() calls Samba does. Jeremy. (This used to be commit d0e48a2d8072c3e77a57ac6a2fb5044c05f03b41)
1998-08-21nmbd and smbd had different behavior with respect to log files. nmbd wouldChristopher R. Hertel1-7/+11
default to overwrite and smbd would default to append. Also, the -a option (actually a toggle, such that "-a -a" would set the default) was documented as append mode for nmbd, and *overwrite mode* for smbd. nmbd now defaults to append mode, to match smbd. The -a option now always means append, and I've added the -o option to both, meaning overwrite. Note that the change to nmbd's default behavior may confuse some people. I've not seen anything about 2.0.0 changes in the WHATSNEW.txt file. Where would I document a change like this? Chris -)----- (This used to be commit b1d374fb14b1fb92a84260f1dcc59a39a4b99a3d)
1998-08-17removed some of the rough edges from the splitupAndrew Tridgell1-435/+451
(This used to be commit 722312c326d54849e2552327fcd88ac77257ad2c)
1998-08-17move soem variables from server.c that don't belong there.Andrew Tridgell1-45/+0
(This used to be commit f92475aa3cb3ade576c39c02c7996c949c42082a)
1998-08-17this completes the splitup of server.c.Andrew Tridgell1-3322/+2
the splitup was done with an axe, not a scalpel, so there are some rough edges. I mostly wanted to get the general form right with fine tuning of what goes where to come later. Still, this is better than what we had before where server.c was a general repository for anything that didn't fit elsewhere. (This used to be commit a6d194886a4a5f7507fa37289ff96c1be56f14a6)
1998-08-17more splitting of server.cAndrew Tridgell1-557/+16
created dosmode.c and filename.c (This used to be commit 534a90ca44641417c21f6ed6d4b94b1de60f808d)
1998-08-17now that we have no global arrays we can start to split up the monsterAndrew Tridgell1-485/+0
server.c without breaking things. this splits off netprot.c and fileio.c for negprot and read/write/seek handling respectively. (This used to be commit b3d7014643ec9f2eef6e6f598f5b9db1fe2f930d)
1998-08-17moved connection_struct handling code into smbd/conn.c and changed itAndrew Tridgell1-119/+18
to a linked list with bitmap format. (This used to be commit b7aaab1b6b2d2f72b2bb7c11f5c7bf081a6093d9)
1998-08-17much cleaner chain pointer handling for both files and pipes.Andrew Tridgell1-4/+1
the chain pointer is now stored as a static and is set whenever a handle is created or extracted. This also makes the code less error prone. (This used to be commit 068a862982bea726e8d7b1b4065d510b9840a272)
1998-08-17some cleanups from the conversion of Pipes[] to a linked list. I alsoAndrew Tridgell1-1/+1
removed most cases where a pnum is used and substituted a pipes_struct*. in files.c I added a offset of 0x1000 to all file handles on the wire. This makes it much less likely that bad parsing will give us the wrong field. (This used to be commit 8bc2627ff28d340db65bfa017daca2dc291d5ef7)
1998-08-16got rid of the Files[] array completely (previously I'd just made itAndrew Tridgell1-2/+2
private to files.c) It now is a doubly linked list with a bitmap for allocated file numbers. Similarly for the fd_ptr code. I also changed the default maximum number of open files to 4096. The static cost is 1 bit per file. It all seems to work, and it passes the "does Sue scream" test, but if you see weird behaviour then please investigate. With the volume of new code that has gone in there are bound to be one or two bugs lurking. note that you must do a "make clean" before building this as many data structures have changed in size. (This used to be commit 79755ce97004b787d7e83a8d18fc4c7c003b7231)
1998-08-16server.c: fixed a bug in close_file() with the new files.c handling codeAndrew Tridgell1-15/+14
bitmap.c: added bitmap hanlding code in preparation for increasing the default max open files to several thousand (This used to be commit f573a65b67e7a57586fec57845598e49b157ee0a)
1998-08-15this checkin gets rid of the global Files[] array and makes it localAndrew Tridgell1-297/+81
in files.c it should now be faily easy to expand the default MAX_OPEN_FILES to many thousands. (This used to be commit b088c804f98908eb02f05ab2f2e8a61691a0a582)
1998-08-15configure: Changes for extra headers.Jeremy Allison1-9/+7
configure.in: Source for header changes. client/clitar.c: Fixed isXXX macros & debugs for gcc pedantic compile. include/config.h.in: Added MEMSET, BZERO, MEMORY, RPCSVC_YPCLNT, STRINGS headers. include/includes.h: Headers for the above. include/smb.h: Made SIGNAL_CAST POSIX by default void (*)(int). lib/access.c: Fixed isXXX macros & debugs for gcc pedantic compile. lib/charset.c: Fixed isXXX macros & debugs for gcc pedantic compile. lib/debug.c: Fixed signal functs. lib/kanji.c: Fixed isXXX macros & debugs for gcc pedantic compile. lib/smbrun.c: Fixed isXXX macros & debugs for gcc pedantic compile. lib/util.c: Fixed isXXX macros & debugs for gcc pedantic compile. libsmb/namequery.c: Fixed isXXX macros & debugs for gcc pedantic compile. locking/shmem.c: Fixed isXXX macros & debugs for gcc pedantic compile. locking/shmem_sysv.c: Fixed error messages in sysV stuff. nmbd/asyncdns.c: Fixed signal functs. nmbd/nmbd.c: Fixed isXXX macros & debugs for gcc pedantic compile. passdb/passdb.c: Fixed isXXX macros & debugs for gcc pedantic compile. passdb/smbpassfile.c: Fixed isXXX macros & debugs for gcc pedantic compile. smbd/chgpasswd.c: Fixed isXXX macros & debugs for gcc pedantic compile. smbd/ipc.c: Fixed isXXX macros & debugs for gcc pedantic compile. smbd/nttrans.c: Fixed fsp code path. smbd/password.c: fixed HAVE_YP_GET_DEFAULT_DOMAIN problem. smbd/printing.c: Fixed isXXX macros & debugs for gcc pedantic compile. smbd/reply.c: Fixed isXXX macros & debugs for gcc pedantic compile. smbd/server.c: Fixed isXXX macros & debugs for gcc pedantic compile. smbd/trans2.c: Fixed core dump bug. smbd/uid.c: Fixed isXXX macros & debugs for gcc pedantic compile. Jeremy. (This used to be commit 1b9cbcd02e575dc0a95fa589f720df30a4acc46b)
1998-08-14this is the bug change to using connection_struct* instead of cnum.Andrew Tridgell1-775/+726
Connections[] is now a local array in server.c I might have broken something with this change. In particular the oplock code is suspect and some .dll files aren't being oplocked when I expected them to be. I'll look at it after I've got some sleep. (This used to be commit c7ee025ead4a85b6fa44a832047b878451845fb6)