[Samba] General Samba Questions
Hi, We've been having lots of issues with our Linux based Samba servers since the Windows domains have migrated to AD. We were hoping and expecting that, at least in the short term, we could run in "mixed" mode and not have to make any changes to our Samba servers. However, things just aren't working well. Also, I've posted several issues to this list over the last several weeks and many of the issues I've encountered have gone unresolved. So, the question(s) I have is what is the recommended/suggested Samba version and configuration we should consider deploying in an infrastructure running with Windows 2003 servers and AD? We are running primarily RedHat 9 and RedHat ES 30 and a majority of our Samba servers are currently running 3.0.7 with some running 2.2.7a, (both of which are RedHat's distributions). We've had all kinds of problems varying from intermittent "password server not available" issues, to smbd locking up and most recently having problems changing a server from server to domain security style. Interestingly, (or maybe not), none of these problems existed prior to the AD upgrades I'm considering making an effort to go full ads mode on the samba servers, however, I've also seen that others have had issues doing this. I'm open to suggestions. Thanks, -John -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/listinfo/samba
[Samba] session setup failed: NT_STATUS_LOGON_FAILURE
Hi, I'm having trouble joining a Samba 2.2.7a server running on RH9 into a domain that is being served by a Windows 2003 server running AD in mixed mode. The server successfully joins the domain when using smbpasswd, however, when running smbclient -L localhost I get the following error: session setup failed: NT_STATUS_LOGON_FAILURE Also when attempting to access shares from windows clients I'm prompted for a username and password, however, no accounts ever get granted access. We have another samba server with the same config, (RH9 and same samba version), and it is working fine. I'm including my smb.conf file below. Has anyone else run into this problem? -John - [global] workgroup = BEDFORD netbios name = AMLINK security = domain encrypt passwords = yes password server = bedford03-dc.atb.teradyne.com wins server = 131.101.77.31 server string = Linux ClearCase Server %v %h log file = /var/log/samba/%m.log log level = 4 max log size = 4000 username map = /etc/samba/smbusers read raw = no oplocks = no kernel oplocks = no level2 oplocks = no create mask = 0774 directory mask = 0775 map archive = No preserve case = yes dont descend = /view deadtime = 0 hide dot files = no [vobs] path = /export/vobs browsable = yes writable = yes public = yes comment = "Vob Storage" [hwvobs] path = /export/vobs/hw_vobs browsable = yes writable = yes public = yes comment = "HWEngVobs" [triggers] path = /var/adm/atria/triggers browsable = yes writable = no write list = vobadm force group = vobgrp force create mode = 664 force directory mode = 775 public = yes comment = "storage for clearcase triggers" csc policy = disable [profiles] path = /var/adm/atria/profiles browsable = yes writable = no write list = vobadm force group = vobgrp force create mode = 664 force directory mode = 775 public = yes comment = "storage for clearcase profiles" csc policy = disable -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/listinfo/samba
[Samba] Re: smbd blocking in receive_smb (how to trigger procedure - LONG)
Hi, We've been experiencing this same problem, however, with Samba 3.0.7-1.3E.1 running on RedHat ES 3.0. We've also experienced this on 2.2.7x on RedHat 9. The problem has been intermittently happening over the last several weeks, (I've posted list about it). This message, along with the others in the thread, describes the symptoms that we're seeing, however, I haven't been able to determine whether this has been fixed or if there is a fix. There were responses in the thread that indicated using the "dead time" parameter to kill off dead processes. Does anyone know if this has been fixed or have any idea whether tunning the "dead time" parameter or some other thing has resolved this? Any help would be appreciated. -John List: samba-technical Subject:Re: smbd blocking in receive_smb (how to trigger procedure - LONG) From: Pascal Date: 2002-08-06 18:29:39 [Download message RAW] I think I finally found a BUG (?) in samba 2.2.5 and now I found a means to reproduce that kind of problem without incriminating any network problem at all (I guess...). But I need a FIFO file on Unix in order to trigger it very easilly. I'd like to know if other (especially SAMBA DEVELOPPERs) can reproduce it. So this is the procedure : 1. take a samba server 2.2.5 (I have mandrake cooker samba-server-2.2.5-14mdk) on a linux kernel version 2.4.18-23mdk 2. samba config : # Global parameters [global] workgroup = VMF netbios name = MYNAME server string = VMF Samba Server %v on %h interfaces = xx.xx.xx.xx/24 yy.yy.yy.yy/24 bind interfaces only = Yes encrypt passwords = Yes map to guest = Bad User syslog = 10 log file = /data/reports/xxx/smblog-%m.txt max log size = 500 name resolve order = wins host bcast unix extensions = Yes socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192 SO_KEEPALIVE os level = 65 lm announce = False preferred master = True domain master = True wins support = Yes message command = /usr/bin/mail -s 'winpopup msg from %f on %m' root < %s; rm %s time offset = 60 panic action = echo samba crash | mail -s 'samba crash' root ;admin users = pascal ivan printer admin = pascal ivan hosts allow = xx.xx.xx.xx/255.255.255.0 yy.yy.yy.yy/255.255.255.0 printing = cups # added by PC 23/07/2002 kernel oplocks = no oplock break wait time = 100 # added by PC 05/08/2002 deadtime = 15 [print$] path = /data/home/smbprinters write list = root pascal ivan [printers] comment = All Printers path = /var/spool/samba create mask = 0700 guest ok = Yes printable = Yes writable = No print command = lpr-cups -P %p -o raw %s -r # using client side printer drivers. lpq command = lpstat -o %p lprm command = cancel %p-%j browseable = No [Docs] comment = Documents VMF path = /data/docpasc valid users = pascal ivan force user = docs force group = docs read only = No create mask = 0664 force create mode = 0664 directory mask = 0750 3. I create a unix FIFO on /data/docpasc mkfifo /data/docpasc/FIFO prw-rw-r--1 pascal docs0 Aug 6 19:26 /data/docpasc/FIFO| 4. take a windows client (win2k, win98se tested) go to the explorer and open the Docs share now click on the fifo name and right-clic to get the menu Your client is frozen 5. go to samba server and try to figure out what happens, here is the log : # smbstatus Samba version 2.2.5 Service uid gid pid machine -- pascal pascal pascal2675 ivpcvmwin98se (192.168.0.126) Tue Aug 6 18:19:45 2002 unsaved pascal vmf-staff 2675 ivpcvmwin98se (192.168.0.126) Tue Aug 6 18:19:46 2002 Docs docs docs 2675 ivpcvmwin98se (192.168.0.126) Tue Aug 6 18:19:45 2002 No locked files # tdbtool is used to check I can read the entire lockfile # ./tdbtool /var/cache/samba/locking.tdb tdb> list freelist: tdb> quit # # client will now open explorer and look at the properties of a FIFO (unix) File on the Docs share # now client is frozen # smbstatus Samba version 2.2.5 Service uid gid pid machine -- IPC$ nobody 65534 2699 pcpcwin (212.208.17.83) Tue Aug 6 18:21:51 2002 pascal pascal pascal2675 ivpcvmwin98se (192.168.0.126) Tue Aug 6 18:19:45 2002 IPC$ nobody 65534 2699 pcpcwin (212.208.17.83) Tue Aug 6 18:21:51 2002 unsaved pascal vmf-staff 2675 ivpcvmwin98se (192.168.0.126) Tue Aug 6 18:19:46 2002 Docs docs docs 2675 ivpcvmwin98se (192.168.0.126) Tue Aug 6 18:19:45 2002 Loc
Re: [Samba] Connection reset by peer
Hi Tom, This particular server is running samba 3.0.7-1.3E.1 on RH 3.0. The domain is running AD in mixed mode, (I'll include the smb.conf below), and I believe the Windows server is Windows 2003. Our clients are a mix of Win2k and WinXp. The server is running Clearcase and the clients access the Vobs via Samba. They are primarily running snapshot views. We have had several issues come up over the last few weeks, (coincidentally when the domain was migrated to Windows AD). The most recent are intermittent problems when "password server not available" shows up in the logs and these particular errors show up, (connection reset by peer... etc...). We recently received a report from a user that he/she will get incomplete snapshot view updates, and/or the update operation itself runs very slowly. During the incomplete view updates the user will get messages like "can't open input file" from Clearcase. We're thinking that there is some sort of authentication problem happening, however, we haven't been able to determine what has been causing all of these problems which are mainly intermittent. I'm hoping that by understanding what is causing these connections errors we'll get to root cause on the others as well. One thing to note, we are also running Samba 2.2.7a on a few RedHat 9 servers which are also running Clearcase. These servers are experiencing the same exact issues and are getting the same errors... These machines have been running with this config for over 1 year and have not had any problems until recently. [global] workgroup = ICD netbios name = VALHALLA security = domain password server = seneca wins server = 131.101.17.18 131.101.17.61 server string = Linux ClearCase Server %v %h log file = /var/log/samba/%m.log log level = 6 max log size = 4000 username map = /etc/samba/smbusers read raw = no oplocks = no kernel oplocks = no level2 oplocks = no create mask = 0774 directory mask = 0775 map archive = No preserve case = yes dont descend = /view deadtime = 0 strict locking = no domain master = no preferred master = no [vobs] path = /export/vobs browsable = yes writable = yes public = yes comment = "Vobs" csc policy = disable "Thomas M. Skeren III" <[EMAIL PROTECTED]> 12/23/2004 08:39 PM To: [EMAIL PROTECTED] cc: [EMAIL PROTECTED] Subject:Re: [Samba] Connection reset by peer [EMAIL PROTECTED] wrote: What are you doing? What configuration. What network configuration? >Hi, > >We continuously see these errors in our logs. I've done some searches on >this but nothing seems to point to anything specific. > >Any ideas? > >-John > > >Dec 23 20:14:41 valhalla smbd[3574]: [2004/12/23 20:14:41, 0] >lib/util_sock.c:get_peer_addr(1000) >Dec 23 20:14:41 valhalla smbd[3574]: getpeername failed. Error was >Transport endpoint is not connected >Dec 23 20:14:41 valhalla smbd[3574]: [2004/12/23 20:14:41, 0] >lib/util_sock.c:get_peer_addr(1000) >Dec 23 20:14:41 valhalla smbd[3574]: getpeername failed. Error was >Transport endpoint is not connected >Dec 23 20:14:41 valhalla smbd[3574]: [2004/12/23 20:14:41, 0] >lib/util_sock.c:write_socket_data(430) >Dec 23 20:14:41 valhalla smbd[3574]: write_socket_data: write failure. >Error = Connection reset by peer >Dec 23 20:14:41 valhalla smbd[3574]: [2004/12/23 20:14:41, 0] >lib/util_sock.c:write_socket(455) >Dec 23 20:14:41 valhalla smbd[3574]: write_socket: Error writing 4 bytes >to socket 5: ERRNO = Connection reset by peer > > -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/listinfo/samba
[Samba] Connection reset by peer
Hi, We continuously see these errors in our logs. I've done some searches on this but nothing seems to point to anything specific. Any ideas? -John Dec 23 20:14:41 valhalla smbd[3574]: [2004/12/23 20:14:41, 0] lib/util_sock.c:get_peer_addr(1000) Dec 23 20:14:41 valhalla smbd[3574]: getpeername failed. Error was Transport endpoint is not connected Dec 23 20:14:41 valhalla smbd[3574]: [2004/12/23 20:14:41, 0] lib/util_sock.c:get_peer_addr(1000) Dec 23 20:14:41 valhalla smbd[3574]: getpeername failed. Error was Transport endpoint is not connected Dec 23 20:14:41 valhalla smbd[3574]: [2004/12/23 20:14:41, 0] lib/util_sock.c:write_socket_data(430) Dec 23 20:14:41 valhalla smbd[3574]: write_socket_data: write failure. Error = Connection reset by peer Dec 23 20:14:41 valhalla smbd[3574]: [2004/12/23 20:14:41, 0] lib/util_sock.c:write_socket(455) Dec 23 20:14:41 valhalla smbd[3574]: write_socket: Error writing 4 bytes to socket 5: ERRNO = Connection reset by peer -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/listinfo/samba
[Samba] Re: Domain password server not available
We continue to see the "password server not available" messages on our servers. I'm including a portion of one of the Samba client logs below. Has anyone else seen anything like this? -John lifelaw2.log:[2004/12/20 14:22:55, 0] smbd/password.c:connect_to_domain_password_server(1290) lifelaw2.log: connect_to_password_server: machine GALA rejected the NetBIOS session request. Error was : Call timed out: server did not respond after 2 milliseconds. lifelaw2.log:[2004/12/20 14:23:16, 0] smbd/password.c:connect_to_domain_password_server(1300) lifelaw2.log: connect_to_domain_password_server: machine OZONE rejected the negotiate protocol. Error was : Call timed out: server did not respond after 2 milliseconds. lifelaw2.log:[2004/12/20 14:23:36, 0] smbd/password.c:connect_to_domain_password_server(1300) lifelaw2.log: connect_to_domain_password_server: machine DYNAMO rejected the negotiate protocol. Error was : Call timed out: server did not respond after 2 milliseconds. lifelaw2.log:[2004/12/20 14:24:17, 0] smbd/password.c:connect_to_domain_password_server(1290) lifelaw2.log: connect_to_password_server: machine GALA rejected the NetBIOS session request. Error was : Call timed out: server did not respond after 2 milliseconds. lifelaw2.log:[2004/12/20 14:24:59, 0] smbd/password.c:connect_to_domain_password_server(1290) lifelaw2.log: connect_to_password_server: machine GALA rejected the NetBIOS session request. Error was : Call timed out: server did not respond after 2 milliseconds. John DeBella 12/20/2004 09:28 AM To: [EMAIL PROTECTED] cc: Subject:Domain password server not available We are running samba 3.0.7-1.3E.1in an AD domain which is running a PDC emulator. The samba servers are configured with "security = domain" and on some servers the password server is specified while on others we are using *. We very frequently see the following messages in the log files of the servers. "domain_client_validate: Domain password server not available" Typically we can count on an smbd process to panic at the same time. We've spoken to the admins of the PDC emulators and they claim that the servers are ok. Has anyone else seen this? Any ideas what would cause this? -John -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/listinfo/samba
[Samba] Domain password server not available
We are running samba 3.0.7-1.3E.1in an AD domain which is running a PDC emulator. The samba servers are configured with "security = domain" and on some servers the password server is specified while on others we are using *. We very frequently see the following messages in the log files of the servers. "domain_client_validate: Domain password server not available" Typically we can count on an smbd process to panic at the same time. We've spoken to the admins of the PDC emulators and they claim that the servers are ok. Has anyone else seen this? Any ideas what would cause this? -John -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/listinfo/samba
[Samba] Changing the password server in smb.conf
Hi, Is it enough to run "service smb reload" when changing the password server setting in smb.conf or is a complete restart required? -John -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/listinfo/samba
[Samba] Errors talking to domain controller
Would anyone know what these errors mean? Dec 14 01:22:36 valhalla smbd[28506]: tdb_chainlock_with_timeout_internal: alarm (10) timed out for key SENECA in tdb /etc/samba/secrets.tdb Dec 14 01:33:33 valhalla smbd[7266]: cli_nt_session_open: cli_nt_create failed on pipe \NETLOGON to machine SENECA. Error was Call returned zero bytes (EOF) Dec 14 01:33:33 valhalla smbd[7266]: connect_to_domain_password_server: unable to open the domain client session to machine SENECA. Error was : Call returned zero bytes (EOF). Dec 14 01:33:33 valhalla smbd[7266]: cli_nt_session_open: cli_nt_create failed on pipe \NETLOGON to machine SENECA. Error was Call returned zero bytes (EOF) Dec 14 01:33:33 valhalla smbd[7266]: connect_to_domain_password_server: unable to open the domain client session to machine SENECA. Error was : Call returned zero bytes (EOF). Dec 14 01:33:33 valhalla smbd[7266]: cli_nt_session_open: cli_nt_create failed on pipe \NETLOGON to machine SENECA. Error was Call returned zero bytes (EOF) Dec 14 01:33:33 valhalla smbd[7266]: connect_to_domain_password_server: unable to open the domain client session to machine SENECA. Error was : Call returned zero bytes (EOF). Dec 14 02:49:43 valhalla smbd[11249]: tdb_chainlock_with_timeout_internal: alarm (10) timed out for key SENECA in tdb /etc/samba/secrets.tdb Dec 14 02:49:43 valhalla smbd[11250]: tdb_chainlock_with_timeout_internal: alarm (10) timed out for key SENECA in tdb /etc/samba/secrets.tdb Dec 14 02:50:49 valhalla smbd[11289]: tdb_chainlock_with_timeout_internal: alarm (10) timed out for key SENECA in tdb /etc/samba/secrets.tdb Dec 14 02:51:14 valhalla smbd[7838]: tdb_chainlock_with_timeout_internal: alarm (10) timed out for key SENECA in tdb /etc/samba/secrets.tdb Dec 14 02:51:22 valhalla smbd[11332]: tdb_chainlock_with_timeout_internal: alarm (10) timed out for key SENECA in tdb /etc/samba/secrets.tdb Dec 14 02:52:47 valhalla smbd[11346]: tdb_chainlock_with_timeout_internal: alarm (10) timed out for key SENECA in tdb /etc/samba/secrets.tdb Dec 14 02:52:54 valhalla smbd[11348]: tdb_chainlock_with_timeout_internal: alarm (10) timed out for key SENECA in tdb /etc/samba/secrets.tdb Dec 14 02:52:57 valhalla smbd[7838]: tdb_chainlock_with_timeout_internal: alarm (10) timed out for key SENECA in tdb /etc/samba/secrets.tdb Dec 14 02:52:57 valhalla smbd[11346]: tdb_chainlock_with_timeout_internal: alarm (10) timed out for key SENECA in tdb /etc/samba/secrets.tdb Dec 14 04:20:25 valhalla nmbd[28500]: ^IICD(1) current master browser = SENECA Dec 14 04:20:25 valhalla nmbd[28500]: ^I^ISENECA 4086102b () -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/listinfo/samba
Re: [Samba] smbd hung processes - Samba 3.0.7
We're still experiencing this issue. I've observed a couple of things during the latest event. I mentioned before that netstat -a shows many smb processes in CLOSE_WAIT state when this problem occurs. I happened to strace on a process that was stuck in "recvfrom(1315,". When I killed this process many of the hung smb processes terminated and the Samba service was responding again. I've noticed the following error in the messages file: Dec 13 18:16:20 valhalla smbd[18005]: [2004/12/13 18:16:20, 0] tdb/tdbutil.c:tdb_log(725) Dec 13 18:16:20 valhalla smbd[18005]: tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 99 ltype =0 (Resource deadlock avoided) This particular error showed up prior to samba being restarted. When the service is restarted many of these show up in the log. I also notice that many times during an event the messages file contains the following errors: Dec 13 19:29:33 valhalla smbd[28820]: domain_client_validate: Domain password server not available. Dec 13 19:29:33 valhalla smbd[28820]: [2004/12/13 19:29:33, 0] auth/auth_domain.c:domain_client_validate(17 0) Not sure if this is a coincidence. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/listinfo/samba
Re: [Samba] smbd hung processes - Samba 3.0.7
This time things behaved a little differently. Samba was not replying to mount requests. An strace of the parent smbd process showed that the requests coming in were attempted to be spawned, however, were failing. I observed this behavior when we had processes hanging and the spawned processes appeared to be hung on a lock. We've since removed strick locking and now the spawned processes die immediately after being spawned. clone(child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0xb72142e8) = 29287 close(22) = 0 select(1024, [18 19 20], NULL, NULL, NULL) = ? ERESTARTNOHAND (To be restarted) --- SIGCHLD (Child exited) @ 0 (0) --- waitpid(-1, NULL, WNOHANG) = 29284 waitpid(-1, NULL, WNOHANG) = 0 sigreturn() = ? (mask now [FPE USR2 PIPE]) select(1024, [18 19 20], NULL, NULL, NULL) = ? ERESTARTNOHAND (To be restarted) --- SIGCHLD (Child exited) @ 0 (0) --- waitpid(-1, NULL, WNOHANG) = 29285 waitpid(-1, NULL, WNOHANG) = 0 sigreturn() = ? (mask now [FPE USR2 PIPE]) select(1024, [18 19 20], NULL, NULL, NULL) = 1 (in [18]) time(NULL) = 1102529684 accept(18, {sa_family=AF_INET, sin_port=htons(1865), sin_addr=inet_addr("131.101.53.201")}, [16]) = 22 fcntl64(22, F_GETFL)= 0x2 (flags O_RDWR) fcntl64(22, F_SETFL, O_RDWR)= 0 clone(child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0xb72142e8) = 29288 close(22) = 0 select(1024, [18 19 20], NULL, NULL, NULL) = ? ERESTARTNOHAND (To be restarted) --- SIGCHLD (Child exited) @ 0 (0) --- waitpid(-1, NULL, WNOHANG) = 29286 waitpid(-1, NULL, WNOHANG) = 0 sigreturn() = ? (mask now [FPE USR2 PIPE]) select(1024, [18 19 20], NULL, NULL, NULL) = 1 (in [18]) time(NULL) = 1102529684 accept(18, {sa_family=AF_INET, sin_port=htons(1304), sin_addr=inet_addr("131.101.18.20")}, [16]) = 22 fcntl64(22, F_GETFL)= 0x2 (flags O_RDWR) fcntl64(22, F_SETFL, O_RDWR)= 0 clone(child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0xb72142e8) = 29289 close(22) = 0 select(1024, [18 19 20], NULL, NULL, NULL) = ? ERESTARTNOHAND (To be restarted) --- SIGCHLD (Child exited) @ 0 (0) --- waitpid(-1, NULL, WNOHANG) = 29287 waitpid(-1, NULL, WNOHANG) = 0 sigreturn() = ? (mask now [FPE USR2 PIPE]) select(1024, [18 19 20], NULL, NULL, NULL) = 1 (in [18]) time(NULL) = 1102529684 accept(18, {sa_family=AF_INET, sin_port=htons(2064), sin_addr=inet_addr("131.101.185.75")}, [16]) = 22 fcntl64(22, F_GETFL)= 0x2 (flags O_RDWR) fcntl64(22, F_SETFL, O_RDWR)= 0 clone(child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0xb72142e8) = 29290 close(22) = 0 select(1024, [18 19 20], NULL, NULL, NULL) = ? ERESTARTNOHAND (To be restarted) --- SIGCHLD (Child exited) @ 0 (0) --- "Gerald (Jerry) Carter" <[EMAIL PROTECTED]> 12/08/2004 11:53 AM To: [EMAIL PROTECTED] cc: [EMAIL PROTECTED] Subject:Re: [Samba] smbd hung processes - Samba 3.0.7 -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 [EMAIL PROTECTED] wrote: | Hmmm. So do you think turning off strict locking will | help or is there something "wrong" with the tdb records | that we can clear? First we need to find out what file that fd is associated with. Then we can start working backwards to find root cause. cheers, jerry - - Alleviating the pain of Windows(tm) --- http://www.samba.org GnuPG Key- http://www.plainjoe.org/gpg_public.asc "If we're adding to the noise, turn off this song"--Switchfoot (2003) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.4 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFBtzGPIR7qMdg1EfYRAi6dAJ9ShhAuixBiW4PLkq2BbM0h7IIF+QCfZjJX Z2Mc3N+SCOQm3RgKfDEwxCY= =ZDqD -END PGP SIGNATURE- -- To unsubscribe from this list go to the following URL and read the instructions: http://lists.samba.org/mailman/listinfo/samba
Re: [Samba] smbd hung processes - Samba 3.0.7
Looks like it's a link to /var/cache/samba/gencache.tdb. -John "Gerald (Jerry) Carter" <[EMAIL PROTECTED]> 12/08/2004 11:53 AM To: [EMAIL PROTECTED] cc: [EMAIL PROTECTED] Subject:Re: [Samba] smbd hung processes - Samba 3.0.7 -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 [EMAIL PROTECTED] wrote: | Hmmm. So do you think turning off strict locking will | help or is there something "wrong" with the tdb records | that we can clear? First we need to find out what file that fd is associated with. Then we can start working backwards to find root cause. cheers, jerry - - Alleviating the pain of Windows(tm) --- http://www.samba.org GnuPG Key- http://www.plainjoe.org/gpg_public.asc "If we're adding to the noise, turn off this song"--Switchfoot (2003) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.4 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFBtzGPIR7qMdg1EfYRAi6dAJ9ShhAuixBiW4PLkq2BbM0h7IIF+QCfZjJX Z2Mc3N+SCOQm3RgKfDEwxCY= =ZDqD -END PGP SIGNATURE- -- To unsubscribe from this list go to the following URL and read the instructions: http://lists.samba.org/mailman/listinfo/samba
Re: [Samba] smbd hung processes - Samba 3.0.7
Hi Jerry, Thanks for the reply. I'll check this if it reoccurs again. We've turned off strict locking to see if this helps. This was on a hunch that it was a lock issue. To answer your question, the access to the main share on this server is via the automounter to a local directory. For example the automount map /hwnet/ccvobs mounts /export/vobs on this server. The share [vobs] is mapped to /hwnet/vobs. The default timeout is 60 seconds and we do see the automounter expire and remount this mount point frequently. While we're not re-exporting this file system there are certainly times when the automounter will apparently unmount and remount it. Note: that during the "event" the filesystem is available both locally and via the automounter. -John "Gerald (Jerry) Carter" <[EMAIL PROTECTED]> 12/08/2004 11:04 AM To: [EMAIL PROTECTED] cc: [EMAIL PROTECTED] Subject:Re: [Samba] smbd hung processes - Samba 3.0.7 -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 [EMAIL PROTECTED] wrote: | We have upgraded to the 3.0.7-1.3E.1 RH Samba update | and this problem still occurs. Has anyone else experienced | this or does anyone have any ideas on what's causing this? | | -John | | | [EMAIL PROTECTED] wrote: | | |> We've seen Samba crash and burn twice in the last 48 hours |> - it just started happening, and we have no idea what |> might be causing it. I'm hoping that someone will |> recognize this problem. Are you reexporting NFS shares by chance? |> in the middle that are not in the smbstatus report. |> What we THINK is happening is that the smbd processes |> begin to hang, the clients time out, A good theory (which would be true if re-exporting NFS shares and the NFS server got stuck). |> # strace -p 20403 |> Process 20403 attached - interrupt to quit |> fcntl64(13, F_SETLKW64, {type=F_RDLCK, whence=SEEK_SET, start=280, |>len=1} |> look in /proc/http://www.samba.org GnuPG Key- http://www.plainjoe.org/gpg_public.asc "If we're adding to the noise, turn off this song"--Switchfoot (2003) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.4 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFBtyX5IR7qMdg1EfYRAmD+AKCvqab8SuxkEFDp8PxPNsqOMJxHmQCfQHpz FMflmk9WH2CP7Jfr52aktkA= =tLj7 -END PGP SIGNATURE- -- To unsubscribe from this list go to the following URL and read the instructions: http://lists.samba.org/mailman/listinfo/samba
Re: [Samba] smbd hung processes - Samba 3.0.7
We have upgraded to the 3.0.7-1.3E.1 RH Samba update and this problem still occurs. Has anyone else experienced this or does anyone have any ideas on what's causing this? -John [EMAIL PROTECTED] wrote: >We've seen Samba crash and burn twice in the last 48 hours - it just >started happening, and we have no idea what might be causing it. I'm >hoping that someone will recognize this problem. > >Platform: we are running RedHat Enterprise Server, with Samba 3.0.7. >We're using security=domain in an old-style NT4 domain environment. > >The symptom that we're seeing is that the number of smbd processes >suddenly begins to increase. We normally run with betwen 100 and 150 smb >processes, but when Samba fails, the number starts to increase quickly, >and users start to have problems accessing files. > >smbstatus reports approximately the right number of clients (133), but ps >shows a much larger number of smbd processes active (680). Smbstatus >reports a list of active smbd processes - this list includes the oldest >processes and the newest processes, but there is a block of smbd processes >in the middle that are not in the smbstatus report. What we THINK is >happening is that the smbd processes begin to hang, the clients time out, >they initiate a new session with Samba server, which respawns another smbd >server process (leaving the old, hung process running). This keeps >happening over and over until we kill samba. The hung processes need to >be kill -9'ed. > >If you do a "strace" on these apparently hung processes, you see this: > ># strace -p 20403 >Process 20403 attached - interrupt to quit >fcntl64(13, F_SETLKW64, {type=F_RDLCK, whence=SEEK_SET, start=280, >len=1} > > >I'm not sure if it's relevent, but netstat -a reports a large number of >sockets in the CLOSE_WAIT state (I've included a small sample): > >Proto Recv-Q Send-Q Local Address Foreign Address State >tcp1 0 valhalla:microsoft-ds army39:1455 CLOSE_WAIT >tcp1 0 valhalla:microsoft-ds 131.101.40.174:2531 >CLOSE_WAIT >tcp 54 0 valhalla:microsoft-ds army39:1435 CLOSE_WAIT >tcp 54 0 valhalla:microsoft-ds 131.101.40.174:2512 >CLOSE_WAIT > >In this log, valhalla is the Samba server, and microsoft-ds is port 445 >(the CIFS port). > >There doesn't seem to be anything relevent in the smbd log files (we were >using log level 1). We've increased the log level to 3 in the hope that >we'll get more information the next time Samba goes wild. > >Our smb.conf file isn't complicated - the global section looks like this: > >[global] > workgroup = ICD > netbios name = VALHALLA > security = domain > password server = * > wins server = nn.nn.nn.nn mm.mm.mm.mm > server string = Linux ClearCase Server %v %h > log file = /var/log/samba/%m.log > log level = 3 > max log size = 4000 > username map = /etc/samba/smbusers > read raw = no > oplocks = no > kernel oplocks = no > level2 oplocks = no > create mask = 0774 > directory mask = 0775 > map archive = No > preserve case = yes > deadtime = 0 > > > Is this by any chance with the 3.0.7-1.3E.1 RH Samba update that was just recently released or one of the previous 3.0.7 RH packages? Christian -- To unsubscribe from this list go to the following URL and read the instructions: http://lists.samba.org/mailman/listinfo/samba