Unable to run process even after running hcidbinit

[Ivan Ng]

I got the following error when starting any of the processes on my site.  I suspect that the DB was corrupted during a power loss.  I tried hcidbinit -A -C and it is still not working.  By the way, what else can be done to recover the information in the db prior to doing hcidbinit?

Thank you in advance.

[prod:prod:INFO/0:  STARTUP_TID] Started at Wed Nov 29 15:14:11 2006

[prod:prod:INFO/0:  STARTUP_TID] Engine process is 4590 on host clover1

11/29/2006 15:14:21

[cmd :cmd :ERR /0:csmicb_o1_cmd] Initialization of DBI failed

11/29/2006 15:14:21

[prod:prod:ERR /0:csmicb_o1_cmd] Unable to initialize the Command Thread.

PANIC: “0”

PANIC: Calling “pti” for thread csmicb_o1_cmd

---

Scheduler State

---

Thread Events     State      Priority Runnable  PT Msgs

  0      0   SCHED_IDLE         0       0       0,0,0

---

Thread 0

---

ti: 0x40160928

   tid           :    0

   HostPthreadId : 0x00000001

   EventList     : 0x40076ac8

   PolledEvents  : 0x40076b10

   PthreadEvent  : 0x4016a898

   ReadyEvents   : 0x40076b28

   CtrlMsgs      : 0x40076b40

   UserCtrlMsgs  : 0x40076b58

   UserDataMsgs  : 0x40076b70

   StartArgs     : 0x00000000

   SchedState    : SCHED_IDLE

   SchedPriority : 0

   Killed        : 0

---

Registered Events

---

el: 0x40076ac8

   elCount : 1

   elHead: 0x40080c70

   elTail: 0x40080c70

ele: 0x40080c70

   event: 0x4016a898

   prev : 0x0

   next : 0x0

ev: 0x4016a898

    evType     : PTHREADS

    evStrDesc    :

    evSocket     : 0

    evMsgQue     : 0

    evTid        : 0

    evState      : 0

    evPtMsg      : 0x0

    evUserData   : 0x0

    evCallBack   : 0x0

    evCbShutdown : 0x0

    evRecurFreq  :

---

Polled Events

---

el: 0x40076b10

   elCount : 0

   elHead: 0x00000000

   elTail: 0x00000000

---

Ready Events

---

el: 0x40076b28

   elCount : 0

   elHead: 0x00000000

   elTail: 0x00000000

---

Outstanding Pthread Ctrl Msgs

---

pmq: 0x40076b40

Count   : 0

Head    : 0x00000000

Tail    : 0x00000000

---

Outstanding Pthread User Ctrl Msgs

---

pmq: 0x40076b58

Count   : 0

Head    : 0x00000000

Tail    : 0x00000000

---

Outstanding Pthread User Data Msgs

---

pmq: 0x40076b70

Count   : 0

Head    : 0x00000000

Tail    : 0x00000000

PANIC: Process panic—engine going down

PANIC: assertion ‘0’ failed at main.cpp/236

Topic

[Russ Ross]

I don’t want to get your hopes up because a corrupted recovery database due to a power loss is not something I know how to recover from without message loss.

If someone knows the answer please share!

However, I would take a look in the directory

$HCISITEDIR/databases; ls -al

to see if there is a left over database lock flag and get rid of it.

I highlighted the lock flag I’m refering to in the attached screen shot.

Note: be sure to do “ls -al” and not “ls -l”.

Russ Ross
RussRoss318@gmail.com

[Robert Kersemakers]

Hi Ivan,

Here’s what we do when we have to restart Cloverleaf:

* Stop all processes

* Stop all site daemons (hcisitectl -f -K)

* Remove the monitorshmemfile (rm $HCISITEDIR/exec/monitorShmemfile)

* Cleanup memory region (hcimsiutil -R)

* Remove vista.taf (rm $HCISITEDIR/exec/databases/vista.taf)

* Cleanup site (hcisitecleanup)

* Initialise database (hcidbinit -iC)

* If necessary (when having DB_vista error), rebuild logs:

   – keybuild rlog

   – dchain rlog

   – keybuild elog

   – dchain elog

* Start site daemons (hcisitectl -S)

* Start all processes

If this fails, we reboot the server. If it still fails, we call support…

Hope this helps.

Zuyderland Medisch Centrum; Heerlen/Sittard; The Netherlands

[Ivan Ng]

Robert, thx, while that does not get the site in problem to work, your list is very useful for future issues.

[Nathan Martin]

Just a thought.  Make sure you’re running as the right user, with the correct permissions on all the directories, and with the correct setroot/showroot settings when you do the dbinit and start the process.

In addition to what Robert said, there may be a command to cleanup after crashes.  That would be:  hcisitecleanup

Also, call the Quovadx support number if you haven’t.

[Author]

Replies