(Illustration by Gaich Muramatsu)
(Illustration by Gaich Muramatsu)
The port mapper seems to be running. Actually, two are. [doug_at_Oak ~]$ ps auxw | grep port bin 326 0.0 0.0 1212 0 ? SW 13:15 0:00 [portmap] root 548 0.0 0.4 2048 304 ? S 13:15 0:00 sendmail: accepting connections on port 25 xfs 615 0.0 1.8 2768 1156 ? S 13:15 0:01 xfs -droppriv -daemon -port -1 root 658 0.0 0.1 1380 104 ? S 13:15 0:00 [rpc2portmap] doug 1349 0.0 0.6 1172 404 pts/1 D 20:30 0:00 grep port [doug_at_Oak ~]$ I've attached as much log data as I could find. After the failed connect, the server seems to still be alive. I can log into venus running on the server and edit files in the coda directory. Doug Ashwin Krishna Patil wrote: > Hey, what did /vice/srv/SrvLog say? Sometimes the server quietly dies but > the process is still visible. Definitely check out the /vice/srv/SrvLog > and /vice/srv/SrvErr on the server side AFTER your failed connect. Also, > make sure that the portmapper is running on the server. > > -------------------------------------------------- > Ashwin Patil > ph. (734) 213-2003 > email: apatil_at_eecs.umich.edu > University of Michigan at Ann Arbor > > On Tue, 13 Jun 2000, Dr. Douglas C. MacKenzie wrote: > > > Thanks for taking an interest. I haven't gotten much farther. > > I did run the bldvldb.sh and makevrdb commands to build the volume > > databases on the server. That seemed to clear up a couple messages, > > but I still can't get the client to successfully start on the other > > machine. > > The client running on the server is happy and mounts /coda fine. > > Both are pentium machines with recent RH6.2 clean loads. > > > > Doug MacKenzie > > doug_at_mobile-intelligence.com > > > > > > The server log /vice/srv/SrvLog has this info: > > > > 3:53:30 File Server started Tue Jun 13 13:53:30 2000 > > > > 13:54:06 client_GetVenusId: got new host 127.0.0.1:2430 > > 13:54:06 Building callback conn. > > 13:54:06 No idle WriteBack conns, building new one > > 13:54:06 Writeback message to 127.0.0.1 port 2430 on conn 7a47eea > > succeeded > > 13:54:06 RevokeWBPermit on conn 7a47eea returned 0 > > 13:54:25 client_GetVenusId: got new host 192.168.1.4:2430 > > 13:54:25 Building callback conn. > > 13:54:25 No idle WriteBack conns, building new one > > 13:54:25 Writeback message to 192.168.1.4 port 2430 on conn 2f911989 > > succeeded > > 13:58:12 VRDB created, 1 entries > > 14:02:15 Callback failed RPC2_DEAD (F) for ws 192.168.1.4, port 2430 > > 14:02:15 Unbinding RPC2 connection 450252675 > > 14:35:17 Building callback conn. > > 14:35:17 RevokeWBPermit on conn 2f911989 returned -2016 > > 14:35:17 No idle WriteBack conns, building new one > > 14:35:17 Writeback message to 192.168.1.4 port 2430 on conn 3f0a69e3 > > succeeded > > 14:42:32 Callback failed RPC2_DEAD (F) for ws 192.168.1.4, port 2430 > > 14:42:32 Unbinding RPC2 connection 40275237 > > > > > > > > The client log /usr/coda/etc/LOG has this info: > > > > [ X(00) : 0000 : 13:33:45 ] Coda Venus, version 5.3.7 > > [ X(00) : 0000 : 13:33:45 ] Logfile initialized with LogLevel = 0 at Tue > > Jun 13 > > 13:33:45 > > 2000 > > > > [ X(00) : 0000 : 13:33:45 ] E StatsInit() > > [ X(00) : 0000 : 13:33:45 ] L StatsInit() > > [ X(00) : 0000 : 13:33:45 ] RecovDataSizes: Log = 0x47f63, Data = > > 0x11fd8c > > [ X(00) : 0000 : 13:33:45 ] Recov_AllocateVM: allocated 1000 bytes at > > 20000000 > > [ X(00) : 0000 : 13:33:45 ] Recov_AllocateVM: allocated 11a000 bytes at > > 21000000 > > [ X(00) : 0000 : 13:33:45 ] BeginRvmTruncate (1, 4312, I) > > [ X(00) : 0000 : 13:33:45 ] cache::EndRvmTruncate > > [ X(00) : 0000 : 13:33:45 ] Recov_DeallocateVM: deallocated 1000 bytes > > at 20000000 > > > > [ X(00) : 0000 : 13:33:45 ] Recov_DeallocateVM: deallocated 11a000 bytes > > at > > 21000000 > > [ X(00) : 0000 : 13:33:45 ] BeginRvmFlush (1, 1332, F) > > [ X(00) : 0000 : 13:33:45 ] EndRvmFlush > > [ X(00) : 0000 : 13:33:45 ] BeginRvmTruncate (2, 1492, F) > > [ X(00) : 0000 : 13:33:45 ] cache::EndRvmTruncate > > [ X(00) : 0000 : 13:33:45 ] BeginRvmFlush (1, 5516, F) > > [ X(00) : 0000 : 13:33:45 ] EndRvmFlush > > [ X(00) : 0000 : 13:33:45 ] BeginRvmTruncate (3, 5676, F) > > [ X(00) : 0000 : 13:33:45 ] cache::EndRvmTruncate > > [ X(00) : 0000 : 13:33:45 ] BeginRvmFlush (1, 223616, F) > > [ X(00) : 0000 : 13:33:45 ] EndRvmFlush > > [ X(00) : 0000 : 13:33:45 ] BeginRvmTruncate (420, 223776, F) > > [ X(00) : 0000 : 13:33:45 ] cache::EndRvmTruncate > > [ X(00) : 0000 : 13:33:45 ] BeginRvmFlush (1, 16808, F) > > [ X(00) : 0000 : 13:33:45 ] EndRvmFlush > > [ X(00) : 0000 : 13:33:45 ] BeginRvmTruncate (1, 16968, F) > > [ X(00) : 0000 : 13:33:45 ] cache::EndRvmTruncate > > [ X(00) : 0000 : 13:33:45 ] E adviceserver::adviceserver: AdviceServer > > > > [ A(18) : 0000 : 13:33:45 ] adviceserver::main() > > > > [ H(07) : 0000 : 13:33:45 ] HDBDaemon about to sleep on hdbdaemon_sync > > > > [ T(01) : 0000 : 13:33:51 ] BeginRvmFlush (1, 2352, T) > > [ T(01) : 0000 : 13:33:51 ] EndRvmFlush > > > > [ D(21) : 0000 : 13:33:51 ] WAITING(SRVRQ): > > > > [ W(20) : 0000 : 13:34:01 ] NotifyUserOfServerDownEvent > > [ W(20) : 0000 : 13:34:01 ] DiscoRef= 1 > > > > [ D(21) : 0000 : 13:34:01 ] WAIT OVER, elapsed = 10023.3 > > > > > > > > > > Jan Harkes wrote: > > > > > On Tue, Jun 13, 2000 at 10:17:35AM -0500, Dr. Douglas C. MacKenzie > > wrote: > > > > I have successfully installed a Coda server and client running on > > RedHat > > > > 6.2. When I tried to start venus on a second machine as just a > > client > > > > I get the following errors in the /var/log/messages > > > ... > > > > Jun 13 10:07:25 maple kernel: Failure of coda_cnode_make for root: > > error -110 > > > > > > ETIMEDOUT, the client was unable to get the attributes of the root > > > directory for the root volume on the server. /usr/coda/venus.log on > > the > > > client and /vice/srv/SrvLog on the server might have more information > > > about why we failed to fetch it. > > > > > > > ----------- > > > > Also, the file /usr/sbin/venus-setup from the rpm file > > > > coda-debug-client-5.3.7-1.i386.rpm > > > > references a utility called codaconfedit > > > > which is not in any of the rpms that I can find. > > > > > > It should be in /usr/sbin, but you are right, it isn't there. I'll > > make > > > sure it ends up in the right place for 5.3.8. > > > > > > Jan > > > > > > ---- client's console messages ----- 08:17:27 /usr/coda/LOG setup for size 0x47f63 08:17:27 /usr/coda/DATA initialized at size 0x11fd8c 08:17:27 brain-wiping recoverable store 08:17:27 loading recoverable store 08:17:27 starting VSGDB scan 08:17:27 0 vsg entries in table 08:17:27 0 vsg entries on free-list 08:17:27 starting VDB scan 08:17:27 1 vol entries in table (0 MLEs) 08:17:27 0 vol entries on free-list (0 MLEs) 08:17:28 starting FSDB scan (416, 10000) (25, 75, 4) 08:17:28 0 cache files in table (0 blocks) 08:17:28 416 cache files on free-list 08:17:28 starting HDB scan 08:17:28 0 hdb entries in table 08:17:28 0 hdb entries on free-list 08:17:28 Kernel version ioctl failed (Inappropriate ioctl for device)! 08:17:28 Initial LRDB allocation 08:17:28 Getting Root Volume information... 08:17:29 Venus starting... [1]+ Killed /usr/sbin/venus -init [root_at_maple etc]# ---------- Client's messages in /var/log/messages -------------------------------- ---------- Is the "coda_downcall: opcode 30, no sb!" message an error? ----------- Jun 14 08:17:27 maple kernel: Coda Kernel/Venus communications (module), v5.0-pre1, braam_at_cs.cmu.edu. Jun 14 08:17:27 maple kernel: coda_psdev_write: downcall, no SB! Jun 14 08:17:29 maple kernel: coda_read_super: rootfid is (0x7f000000,0x1,0x1) Jun 14 08:17:44 maple kernel: coda_downcall: opcode 30, no sb! Jun 14 08:17:45 maple kernel: Failure of coda_cnode_make for root: error -110 ---------- Client's messages in /usr/coda/etc/venus.log --------------- [ X(00) : 0000 : 08:17:27 ] Coda Venus, version 5.3.7 [ X(00) : 0000 : 08:17:27 ] Logfile initialized with LogLevel = 0 at Wed Jun 14 08:17:27 2000 [ X(00) : 0000 : 08:17:27 ] E StatsInit() [ X(00) : 0000 : 08:17:27 ] L StatsInit() [ X(00) : 0000 : 08:17:27 ] RecovDataSizes: Log = 0x47f63, Data = 0x11fd8c [ X(00) : 0000 : 08:17:27 ] Recov_AllocateVM: allocated 1000 bytes at 20000000 [ X(00) : 0000 : 08:17:27 ] Recov_AllocateVM: allocated 11a000 bytes at 21000000 [ X(00) : 0000 : 08:17:27 ] BeginRvmTruncate (1, 4312, I) [ X(00) : 0000 : 08:17:27 ] cache::EndRvmTruncate [ X(00) : 0000 : 08:17:27 ] Recov_DeallocateVM: deallocated 1000 bytes at 20000000 [ X(00) : 0000 : 08:17:27 ] Recov_DeallocateVM: deallocated 11a000 bytes at 21000000 [ X(00) : 0000 : 08:17:27 ] BeginRvmFlush (1, 1332, F) [ X(00) : 0000 : 08:17:27 ] EndRvmFlush [ X(00) : 0000 : 08:17:27 ] BeginRvmTruncate (2, 1492, F) [ X(00) : 0000 : 08:17:27 ] cache::EndRvmTruncate [ X(00) : 0000 : 08:17:27 ] BeginRvmFlush (1, 5516, F) [ X(00) : 0000 : 08:17:27 ] EndRvmFlush [ X(00) : 0000 : 08:17:27 ] BeginRvmTruncate (3, 5676, F) [ X(00) : 0000 : 08:17:27 ] cache::EndRvmTruncate [ X(00) : 0000 : 08:17:28 ] BeginRvmFlush (1, 223616, F) [ X(00) : 0000 : 08:17:28 ] EndRvmFlush [ X(00) : 0000 : 08:17:28 ] BeginRvmTruncate (420, 223776, F) [ X(00) : 0000 : 08:17:28 ] cache::EndRvmTruncate [ X(00) : 0000 : 08:17:28 ] BeginRvmFlush (1, 16808, F) [ X(00) : 0000 : 08:17:28 ] EndRvmFlush [ X(00) : 0000 : 08:17:28 ] BeginRvmTruncate (1, 16968, F) [ X(00) : 0000 : 08:17:28 ] cache::EndRvmTruncate [ X(00) : 0000 : 08:17:28 ] E adviceserver::adviceserver: AdviceServer [ A(18) : 0000 : 08:17:28 ] adviceserver::main() [ H(07) : 0000 : 08:17:28 ] HDBDaemon about to sleep on hdbdaemon_sync [ T(01) : 0000 : 08:17:34 ] BeginRvmFlush (1, 2352, T) [ T(01) : 0000 : 08:17:34 ] EndRvmFlush [ D(21) : 0000 : 08:17:34 ] WAITING(SRVRQ): [ W(20) : 0000 : 08:17:44 ] NotifyUserOfServerDownEvent [ W(20) : 0000 : 08:17:44 ] DiscoRef= 1 [ D(21) : 0000 : 08:17:44 ] WAIT OVER, elapsed = 10016.7 ----------- The RPC port mapper is running on the client --------- ----------- Hmm, but not the RPC2 port mapper??? --------- [root_at_maple etc]# ps axuw | grep -i port bin 332 0.0 0.1 1212 428 ? S 08:11 0:00 portmap root 541 0.0 0.4 2048 1056 ? S 08:11 0:00 sendmail: accepting connections on port 25 xfs 595 0.0 0.5 2264 1372 ? S 08:11 0:00 xfs -droppriv -daemon -port -1 ----------- The Server's messages from /vice/srv/SrvLog ---------- ----------- /vice/srv/SrvErr is empty ---------- 09:12:27 New SrvLog started at Wed Jun 14 09:12:27 2000 09:12:27 Resource limit on data size are set to 2147483647 09:12:27 Server etext 0x80c0c76, edata 0x80ed5b0 09:12:27 RvmType is Rvm 09:12:27 Main process doing a LWP_Init() 09:12:27 Main thread just did a RVM_SET_THREAD_DATA 09:12:27 Setting Rvm Truncate threshhold to 5. Partition /vicepa: inodes in use: 1, total: 1048576. 09:12:40 Partition /vicepa: 280633K available (minfree=5%), 280622K free. 09:12:40 The server (pid 661) can be controlled using volutil commands 09:12:40 "volutil -help" will give you a list of these commands 09:12:40 If desperate, "kill -SIGWINCH 661" will increase debugging level 09:12:40 "kill -SIGUSR2 661" will set debugging level to zero 09:12:40 "kill -9 661" will kill a runaway server 09:12:40 Vice file system salvager, version 3.0. 09:12:40 SanityCheckFreeLists: Checking RVM Vnode Free lists. 09:12:40 DestroyBadVolumes: Checking for destroyed volumes. 09:12:40 Salvaging file system partition /vicepa 09:12:40 Force salvage of all volumes on this partition 09:12:40 Scanning inodes in directory /vicepa... 09:12:40 Entering DCC(0x1000001) 09:12:40 DCC: Salvaging Logs for volume 0x1000001 09:12:40 done: 2 files/dirs, 3 blocks 09:12:40 SalvageFileSys completed on /vicepa 09:12:40 VAttachVolumeById: vol 1000001 (coda_root.0) attached and online 09:12:40 Attached 1 volumes; 0 volumes not attached lqman: Creating LockQueue Manager.....LockQueue Manager starting ..... 09:12:40 LockQueue Manager just did a rvmlib_set_thread_data() done 09:12:40 CallBackCheckLWP just did a rvmlib_set_thread_data() 09:12:40 CheckLWP just did a rvmlib_set_thread_data() 09:12:40 ServerLWP 0 just did a rvmlib_set_thread_data() 09:12:40 ServerLWP 1 just did a rvmlib_set_thread_data() 09:12:40 ServerLWP 2 just did a rvmlib_set_thread_data() 09:12:40 ServerLWP 3 just did a rvmlib_set_thread_data() 09:12:40 ServerLWP 4 just did a rvmlib_set_thread_data() 09:12:40 ServerLWP 5 just did a rvmlib_set_thread_data() 09:12:40 ResLWP-0 just did a rvmlib_set_thread_data() 09:12:40 ResLWP-1 just did a rvmlib_set_thread_data() 09:12:41 VolUtilLWP 0 just did a rvmlib_set_thread_data() 09:12:41 VolUtilLWP 1 just did a rvmlib_set_thread_data() 09:12:41 Starting SmonDaemon timer 09:12:41 File Server started Wed Jun 14 09:12:41 2000 09:12:42 client_GetVenusId: got new host 127.0.0.1:2430 09:12:42 Building callback conn. 09:12:42 No idle WriteBack conns, building new one 09:12:42 Writeback message to 127.0.0.1 port 2430 on conn 3a416ecc succeeded 09:12:42 RevokeWBPermit on conn 3a416ecc returned 0 >>> Here is where I started the remote venus client, the clocks are skewed <<<< 09:18:59 client_GetVenusId: got new host 192.168.1.4:2430 09:19:00 Building callback conn. 09:19:00 No idle WriteBack conns, building new one 09:19:00 Writeback message to 192.168.1.4 port 2430 on conn b036baf succeeded ------------------------------------------------------------------------------ --- The server's /var/log/messages file. Is the "coda_psdev_write: downcall, no SB!" --- message an error? Also, I notice that the client does not start auth2.init in --- the boot process, if that matters. ---------------------------------------- un 14 09:12:26 Oak auth2.init: Starting auth2: Jun 14 09:12:26 Oak auth2.init: /usr/sbin/auth2 done. Jun 14 09:12:26 Oak rc: Starting auth2.init succeeded Jun 14 09:12:26 Oak update.init: Starting coda update servers: Jun 14 09:12:26 Oak update.init: rpc2portmap Jun 14 09:12:26 Oak update.init: updatesrv Jun 14 09:12:26 Oak update.init: updateclnt done. Jun 14 09:12:26 Oak rc: Starting update.init succeeded Jun 14 09:12:26 Oak codasrv.init: Starting codasrv: Jun 14 09:12:26 Oak codasrv.init: codasrv. Jun 14 09:12:27 Oak rc: Starting codasrv.init succeeded Jun 14 09:12:27 Oak venus.init: Starting venus: Jun 14 09:12:27 Oak venus.init: done. Jun 14 09:12:28 Oak rc: Starting venus.init succeeded Jun 14 09:12:28 Oak linuxconf: Linuxconf final setup Jun 14 09:12:28 Oak kernel: Coda Kernel/Venus communications (module), v5.0-pre1, braam_at_cs.cmu.edu. Jun 14 09:12:28 Oak kernel: coda_psdev_write: downcall, no SB! Jun 14 09:12:36 Oak rc: Starting linuxconf succeeded Jun 14 09:12:42 Oak kernel: coda_psdev_write: downcall, no SB! Jun 14 09:12:42 Oak kernel: coda_psdev_write: downcall, no SB! Jun 14 09:12:42 Oak kernel: coda_read_super: rootfid is (0x7f000000,0x1,0x1) Jun 14 09:12:42 Oak kernel: coda_read_super: rootinode is 1025 dev 3 Jun 14 09:14:19 Oak PAM_pwdb[707]: (login) session opened for user doug by LOGIN(uid=0) Jun 14 09:14:42 Oak gnome-name-server[785]: starting Jun 14 09:14:42 Oak gnome-name-server[789]: starting Jun 14 09:14:43 Oak gnome-name-server[785]: name server starting Jun 14 09:14:43 Oak gnome-name-server[789]: name server was running on display, exiting Jun 14 09:15:15 Oak PAM_pwdb[813]: (su) session opened for user root by doug(uid=500) >> Here is where I started the remote venus client <<< >> There were no messages in the Server's /var/log/messages file <<< >> after I started the client. <<<Received on 2000-06-14 09:43:46