Command line hcienginerun doesn’t work

[Tim Jipson]

At the command line I can run: hcienginestop -p p_all  and it shuts down the process just fine.

When I type: hcienginerun -p p_all I get the message: Process p_all initiated but nothing happens. If I type this same command in the gui remote commands area then the process starts just fine.

At the command line setroot and setsite have been run and showroot confirms the settings.

What would cause this?

Thanks,

Tim J.

Topic

[Ron Archambault]

Take a look at your p_all.err, p_all.log and startup_log in the $HCISITEDIR/exec/processes/p_all. This should provide some insight.

[Russ Ross]

Since it is working from the GUI but not the manual backend login, I’m curious if you are logging in as the hci user on the backend or if you or logging in with another user ID.

Russ Ross
RussRoss318@gmail.com

[Tim Jipson]

I am logging in as hci. Also we are on Cloverleaf 5.6PRev2 with Redhat RHEL 5.0

There are errors in the p_all.log but no errors show up when the process is started via the gui.

Code:

[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] Copyright 1993-2008, Quovadx Inc.
[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] CLOVERLEAF(R) Integration Services 5.6P Rev2
[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] Linked by root on host=(Linux enicar.huatek.com 2.6.9-42.ELsmp #1 SMP Wed Jul 12 23:27:17 EDT 2006 i686 i686 i386 GNU/Linux) at Mon Jun  2 15:32:14 2008 in /work/jerickso/cloverrel/cloverleaf/engine/main (build 2)
[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] FEATURE cl-pkg-cl hcilicmgrd 5.6 permanent uncounted 8CFAC25B56EF
[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] HOSTID=001cc4eb05c2 ck=178 category=testdev
[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] FEATURE cl-intfc-thread-50 hcilicmgrd 5.6 permanent uncounted
[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] 99EFD5B926A3 HOSTID=001cc4eb05c2 ck=67 category=testdev
[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] FEATURE cl-mm-master hcilicmgrd 5.6 permanent uncounted D172EBDBF90A
[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] HOSTID=001cc4eb05c2 ck=103 category=testdev
[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] Started at Fri Jun 26 09:48:13 2009
[prod:prod:INFO/0:  STARTUP_TID:06/26/2009 09:48:13] Engine process is 4239 on host nordx-cltest.mmc.org
[dbi :dbi :ERR /0:    p_all_cmd:06/26/2009 09:48:13] Unable to load RDM shared library: Cannot load RIL library libclrdmedb5.6.so:    -1

[cmd :cmd :ERR /0:    p_all_cmd:06/26/2009 09:48:13] Initialization of DBI failed
[prod:prod:ERR /0:    p_all_cmd:06/26/2009 09:48:13] Unable to initialize the Command Thread.
PANIC: “0”
PANIC: Calling “pti” for thread p_all_cmd
—– Scheduler State —–
Thread Events     State      Priority Runnable  PT Msgs
  0      0   SCHED_IDLE         0       0       0,0,0

————– Thread 0 ————–
ti: 0x0x9d7d6e8
   tid           :    0
   HostPthreadId : 0x0xb7f766d0
   EventList     : 0x0x9d7b700
   PolledEvents  : 0x0x9d7d640
   PthreadEvent  : 0x0x9d7d828
   ReadyEvents   : 0x0x9d7d490
   CtrlMsgs      : 0x0x9d7d7e8
   UserCtrlMsgs  : 0x0x9d7d7f8
   UserDataMsgs  : 0x0x9d7d808
   StartArgs     : 0x(nil)
   SchedState    : SCHED_IDLE
   SchedPriority : 0
   Killed        : 0
—– Registered Events —–
el: 0x0x9d7b700
   elCount : 1
   elHead: 0x0x9d7d868
   elTail: 0x0x9d7d868
ele: 0x0x9d7d868
   event: 0x0x9d7d828
   prev : 0x0
   next : 0x0
ev: 0x9d7d828
    evType     : PTHREADS
    evStrDesc    :
    evSocket     : 0
    evMsgQue     : 0
    evTid        : 0
    evState      : 0
    evPtMsg      : 0x0
    evUserData   : 0x0
    evCallBack   : 0x0
    evCbShutdown : 0x0
    evRecurFreq  :
—– Polled Events —–
el: 0x0x9d7d640
   elCount : 0
   elHead: 0x(nil)
   elTail: 0x(nil)
—– Ready Events —–
el: 0x0x9d7d490
   elCount : 0
   elHead: 0x(nil)
   elTail: 0x(nil)
—– Outstanding Pthread Ctrl Msgs —–
pmq: 0x0x9d7d7e8
Count   : 0
Head    : 0x(nil)
Tail    : 0x(nil)
—– Outstanding Pthread User Ctrl Msgs —–
pmq: 0x0x9d7d7f8
Count   : 0
Head    : 0x(nil)
Tail    : 0x(nil)
—– Outstanding Pthread User Data Msgs —–
pmq: 0x0x9d7d808
Count   : 0
Head    : 0x(nil)
Tail    : 0x(nil)

PANIC: Process panic—engine going down
PANIC: assertion ‘0’ failed at main.cpp/254

[Ron Archambault]

try hcilmclear -p p_all

[Tim Jipson]

hcilmclear -p p_all returned no messages of any kind and didn’t change anything.

Also just rebooted, no change.[/code]

[Ron Archambault]

You might have to initialize the DB. This command is destructive so make sure nothing else in your site is running and no messages are in the queue.

hcidbinit

[Rob Abbott]

It looks like your PATH or LIBPATH (or some other environment variable) may be broken.

Are you doing a setroot before executing anything from the command line?

Rob Abbott
Cloverleaf Emeritus

[Tim Jipson]

I don’t seem to have a LIBPATH defined.

Code:

-bash-3.1$ echo $PATH
/opt/quovadx/qdx5.6/integrator/site_cltest/bin:/opt/quovadx/qdx5.6/integrator/site_cltest/scripts:/opt/quovadx/qdx5.6/integrator/bin:/opt/quovadx/qdx5.6/integrator/contrib:/opt/quovadx/qdx5.6/integrator/sbin:/opt/quovadx/qdx5.6/integrator/dbms/bin:/opt/quovadx/qdx5.6/integrator/tcl/bin:/opt/quovadx/qdx5.6/integrator/clgui/bin:/opt/quovadx/qdx5.6/integrator/clgui/java/bin:/opt/quovadx/qdx5.6/integrator/usercmds:/usr/bin:/etc:/usr/sbin:/usr/ucb:/home/hci/bin:/usr/bin/X11:/sbin:.:/usr/local/bin:/usr/local/scripts:/bin
-bash-3.1$ echo $LIBPATH

-bash-3.1$

Can you tell me is this sequence of commands make sense? Everytime I type setroot I go back to defaults.

Code:

-bash-3.1$ showroot
HCI root is /opt/quovadx/qdx5.6/integrator
HCI master site is site_cltest
HCI site is site_cltest
-bash-3.1$ setsite site_dev_tim
-bash-3.1$ showroot
HCI root is /opt/quovadx/qdx5.6/integrator
HCI master site is site_cltest
HCI site is site_dev_tim
-bash-3.1$ setroot
-bash-3.1$ showroot
HCI root is /opt/quovadx/qdx5.6/integrator
HCI master site is site_cltest
HCI site is site_cltest
-bash-3.1$

[Tom Rioux]

After you set your site, you shouldn’t need to set your root again.  As you can see by your showroot after the setsite, the root appears to be correct.  Is this the sequence you are doing before attempting to start the threads back up?  If so, leave off the setroot and you should be fine.  Just make sure you are in the site where you want to start your processes.

Hope this helps…

Tom Rioux

[Tim Jipson]

I just found out that my setroot usage was wrong(I think). When I type: setroot /opt/quovadx/qdx5.6/integrator site_dev_tim running a show root gives what I need. But hcienginerun still doesn’t work.

If I can get this working it will be a part of a backup script and web portal.

Also, from the command line “hcienginestop -p p_all” works fine. I just can’t use hcienginerun.

[Russ Ross]

Grasping for straws again, but I always do a “setroot -clear” prior to any setroot because I had found in a past version of cloverleaf some left over PATHS lived on and caused problems.

For example my scripted setroot steps that are in many of our scripts is as follows:

Russ Ross
RussRoss318@gmail.com

[Russ Ross]

Now that I’m starting to wonder if you might have multiple versions of cloverleaf getting in the way of each other do these steps and let us know.

At the command prompt type functions and post what the setroot function lists.

For Cloverleaf 5.6 mine looks like this:

Code:

function setroot
{
   eval `${CL_INSTALL_DIR}/integrator/sbin/hcisetenv -root ksh $*`
}

For Cloverleaf 5.2 mine looks like this:

Code:

function setroot
{
   eval `${QUOVADX_INSTALL_DIR}/integrator/sbin/hcisetenv -root ksh $*`
}

So for cloverleaf 5.6 it will be necessary to have the environment variable ${CL_INSTALL_DIR} also set so find out by doing

    echo ${CL_INSTALL_DIR}

If it isn’t set or is incorrect then an entry in the /etc/envrionment file something like

    CL_INSTALL_DIR=/quovadx/qdx5.6

will be necessary.

Installing cloverleaf will typically make this entry for you but if you have a fail-over cluster you have to either install cloverleaf after failing over to each cluster or know all the things you have to manually modify, which wouldn’t be my first choice.

Which leads me to another potentially important question, “Is you problematic backend login to a node in a fail-over cluster and was cloverleaf initially install on the problematic node in the fail-over cluster or one of the other nodes?”

Russ Ross
RussRoss318@gmail.com

[Tim Jipson]

setroot -clear doesn’t seem to have changed anything.

I ran the path script that you posted, the only difference(besides my site name) was that I also have “/bin” listed.

$FPATH is displaying correctly, however my environment file is blank. FPATH is being set via /home/hci/.profile

This server was built fresh for 5.6, no other versions are present and we do not use clustering.

I don’t believe that this is related but I’ve recently discovered that hciverify is throwing an error. I sent this issue to cloverleaf support last week but haven’t heard anything yet.

Code:

-bash-3.1$ hciverify
*** Error: uname and hostname don’t match: nordx-cltest != cltest

[Tim Jipson]

FYI… This was casued by installing Apache on the cloverleaf server.

[Russ Ross]

How did installing Apache cause a problem and what did you do to fix it?

I think we have Apache installed on at least one cloverleaf server and don’t think we have the problem you have been experiencing.

Russ Ross
RussRoss318@gmail.com

[David Barr]

Russ Ross wrote:

Here is a script

[Author]

Replies