rpcd(dced) failed to start at cloned system, installed by ignite's goldimage.

[H.Ohkubo]

Hello,

After installed HP-UX11.00(model9000/785/C3600) system from ignite's goldimage archive,
rpcd/dced failed to start on cloned system.
Source system and cloned system are same model and same OS version,
but this phenomenon is not seen on the source system of goldimage archive.

Information about cloned system are

1. swlist information(extract):
B6192AA B.11.00.10 DCE/9000 Programming & Administration Tools Media and Manuals
B6733AA B.11.00.10 DCE/9000 Kernel Threads Support

2. Output messages when try to manual start:
# /sbin/rc2.d/S590Rpcd start
starting the DCE RPC Daemon
/opt/dce/sbin/rpcd
EXIT CODE: 1

3. Try to manual start caused core output under /var/opt/dce/dced
strings of core(extract):

B$Revision: 92453-07 linker linker crt0.o B.10.60 970821 $
/usr/lib/dld.sl
ERROR: mmap failed for dld
ERROR: mmap failed for TSD
@(#)HP DCE/9000 1.7-1.7.1 PHSS_20444-47 Module: dced (Export) Date: Feb 29 2000 18:08:09

Any help would be appreciated!
Thank you, H.Ohkubo

[Ermin Borovac]

Any errors in /var/opt/dce/svc/*.log?

[H.Ohkubo]

Hello, Ermin,
Thanks for your reply!

Yes, there is 3 log files in /var/opt/dce/svc/
1. error.log
2. fatal.log
3. warning.log

Latest update is "error log", and each output is below.

1. error.log:

2005-05-20-06:51:34.705+00:00I----- dced ERROR dhd general main.c 1326 0x7af6f5a0 waitpid() return status 11 (for pid 1095), core dumped
2005-05-20-07:08:11.555+00:00I----- dced ERROR dhd general main.c 1326 0x7af6f5a0 waitpid() return status 11 (for pid 1097), core dumped

Date is right, but I didn't any operation at the time(06:51:34 & 07:08:11)...
System clock uses xntpd and date command returns right date and time.

# date
Fri May 20 16:38:54 JST 2005

2. fatal.log:

2005-05-19-06:17:21.743+00:00I----- dced FATAL dhd general main.c 1521 0x7af6f5a0 Invalid protocol sequence "start".

Also I didn't anything at the time...

3. warning.log:

No output in this file.(size=0)

Any help would be appreciated!
Thank you, H.Ohkubo

[Ermin Borovac · Accepted solution]

Check block size for /var.

# fstyp -v /dev/vg00/rlvolx | grep f_bsize

If it's 64k or more you should install the following patch.

Patch Name: PHSS_29749
Patch Description: s700_800 11.00 HP DCE/9000 1.7 Runtime cumulative patch

1. JAGae41968/JAGae56569 : As the existing implementation of dce daemon (dced/rpcd) does not support a file system of block size greater than 32K, access to the directory /var/opt/dce fails.

Resolution : The code has been modified to handle the file system block sizes greater than 32K.

[H.Ohkubo]

Hello, Ermin,

Thanks and sorry for late reply!

It's solved!

Applied pathes(PHSS_29749) with it's premise patches,
it succeeded in starting of rpcd.

Start RPC daemon if needed
Output from "/sbin/rc2.d/S590Rpcd start":
----------------------------
starting the DCE RPC Daemon
/opt/dce/sbin/rpcd

Thank you! with my respect!

H.Ohkubo

[H.Ohkubo]

Thanks for your time to help me.