Highlighted
Absent Member.
Absent Member.
3719 views

novell-pbserv status dead

Running ZCM 10.3.0 on four primary servers (SLES 10 SP3 64 bit)

Our first server in the zone (that also holds the certificate) often has problems with the service:
novell-pbserv. The services dies aprox. every 14 days. Sometimes once a week. The servers ip is 10.108.128.101. Never seen this problem on other primaries.

If I do a /etc/init.s/novell-pbserv status it reports:
Checking for Novell ZENwWorks Imaging Service: dead
A /etc/init.s/novell-pbserv stop & start allways gets it up and running again.

The novell-pbserv.log reports
Mon Jan 10 14:35:23 2011 -- [ZENIMGWEB-CSCKT] Failed to connect to ZCM server 10.108.128.101 at port 443. Error code: 12. Certificate doesn't verify.
Mon Jan 10 14:35:23 2011 -- [ZIMGWEB-OHC] cannot connect to server : 0
Mon Jan 10 14:35:23 2011 -- [ZENIMGWEB-GPF] Error opening the socket : -602
Mon Jan 10 14:35:23 2011 -- [RPM] PFPS = 22
Mon Jan 10 14:35:23 2011 -- [PCPM] done, PROXY_CMD_PREBOOT_MENU
Mon Jan 10 14:35:23 2011 -- [PCC] received PROXY_CMD_CLOSE
Mon Jan 10 14:35:23 2011 -- [PCC] done PROXY_CMD_CLOSE
Mon Jan 10 14:35:23 2011 -- [LC] close connection, 10.108.128.101


It seems like it can't connect to itself. Is this a known bug? Anyone else seen this error?

TID 7006265 mentions a "Bug 598457 Novell-PBServ Hangs in Mixed ZDM/ZCM/ENGL Environment" but we don't use ENGL nor ZDM.

/Anders
Labels (2)
0 Likes
12 Replies
Highlighted
Knowledge Partner
Knowledge Partner

Re: novell-pbserv status dead

martinusen;2063462 wrote:
Running ZCM 10.3.0 on four primary servers (SLES 10 SP3 64 bit)

Our first server in the zone (that also holds the certificate) often has problems with the service:
novell-pbserv. The services dies aprox. every 14 days. Sometimes once a week. The servers ip is 10.108.128.101. Never seen this problem on other primaries.

If I do a /etc/init.s/novell-pbserv status it reports:
Checking for Novell ZENwWorks Imaging Service: dead
A /etc/init.s/novell-pbserv stop & start allways gets it up and running again.

The novell-pbserv.log reports
Mon Jan 10 14:35:23 2011 -- [ZENIMGWEB-CSCKT] Failed to connect to ZCM server 10.108.128.101 at port 443. Error code: 12. Certificate doesn't verify.
Mon Jan 10 14:35:23 2011 -- [ZIMGWEB-OHC] cannot connect to server : 0
Mon Jan 10 14:35:23 2011 -- [ZENIMGWEB-GPF] Error opening the socket : -602
Mon Jan 10 14:35:23 2011 -- [RPM] PFPS = 22
Mon Jan 10 14:35:23 2011 -- [PCPM] done, PROXY_CMD_PREBOOT_MENU
Mon Jan 10 14:35:23 2011 -- [PCC] received PROXY_CMD_CLOSE
Mon Jan 10 14:35:23 2011 -- [PCC] done PROXY_CMD_CLOSE
Mon Jan 10 14:35:23 2011 -- [LC] close connection, 10.108.128.101


It seems like it can't connect to itself. Is this a known bug? Anyone else seen this error?

TID 7006265 mentions a "Bug 598457 Novell-PBServ Hangs in Mixed ZDM/ZCM/ENGL Environment" but we don't use ENGL nor ZDM.

/Anders


Did you install ZCM with default ports (80/443) or did you change them during the installation?

Thomas
0 Likes
Highlighted
Absent Member.
Absent Member.

Re: novell-pbserv status dead

ZCM is installed on standard ports...

/Anders
0 Likes
Highlighted
Knowledge Partner
Knowledge Partner

Re: novell-pbserv status dead

martinusen;2064229 wrote:
ZCM is installed on standard ports...

/Anders


Can you post the output of /etc/opt/novell/zenworks/preboot/novell-zenimgweb.conf ?

Thomas
0 Likes
Highlighted
Absent Member.
Absent Member.

Re: novell-pbserv status dead

thsundel;2064230 wrote:
Can you post the output of /etc/opt/novell/zenworks/preboot/novell-zenimgweb.conf ?


Novell-zenimgweb.conf:
# Specify the HTTP server IPAddress 
IPADDRESS=10.108.128.101

# HTTP port
HTTP_PORT=80

# HTTPS port
HTTPS_PORT=443

# Allow HTTP if HTTPS is not available
CLEARTEXT=FALSE

# HTTP Service Path
HTTP_PATH=/zenworks-preboot/

# Timer for Cache work to do incase of Satellite Server in milliseconds
CACHETIMER=300000



BTW the service just died yesterday evening. There were no entries in the novell-pbserv.log, but /var/log/messages shows:
Jan 12 23:08:44 zcm01 kernel: novell-pbservd[1412]: segfault at 0000000000000479 rip 00000000f7cd3b6a rsp 00000000f42faf54 error 4


/Anders
0 Likes
Highlighted
Knowledge Partner
Knowledge Partner

Re: novell-pbserv status dead

martinusen;2064252 wrote:
Novell-zenimgweb.conf:
# Specify the HTTP server IPAddress 
IPADDRESS=10.108.128.101

# HTTP port
HTTP_PORT=80

# HTTPS port
HTTPS_PORT=443

# Allow HTTP if HTTPS is not available
CLEARTEXT=FALSE

# HTTP Service Path
HTTP_PATH=/zenworks-preboot/

# Timer for Cache work to do incase of Satellite Server in milliseconds
CACHETIMER=300000



BTW the service just died yesterday evening. There were no entries in the novell-pbserv.log, but /var/log/messages shows:
Jan 12 23:08:44 zcm01 kernel: novell-pbservd[1412]: segfault at 0000000000000479 rip 00000000f7cd3b6a rsp 00000000f42faf54 error 4


/Anders


Our novell-zenimgweb.conf looks like this:

# Specify the HTTP server IPAddress
#IPADDRESS=XXX.XXX.XXX.XXX

# HTTP port
#HTTP_PORT=80

# HTTPS port
#HTTPS_PORT=443

# Allow HTTP if HTTPS is not available
#CLEARTEXT=FALSE

# HTTP Service Path
#HTTP_PATH=/zenworks-preboot/

# Timer for Cache work to do incase of Satellite Server in milliseconds
CACHETIMER=300000

You could try changing it as my conf and see if it helps.. I think it may be the CLEARTEXT=FALSE value that mig be causing this issue..

Thomas
0 Likes
Highlighted
Absent Member.
Absent Member.

Re: novell-pbserv status dead

thsundel;2064254 wrote:
Our novell-zenimgweb.conf looks like this:

# Specify the HTTP server IPAddress
#IPADDRESS=XXX.XXX.XXX.XXX

# HTTP port
#HTTP_PORT=80

# HTTPS port
#HTTPS_PORT=443

# Allow HTTP if HTTPS is not available
#CLEARTEXT=FALSE

# HTTP Service Path
#HTTP_PATH=/zenworks-preboot/

# Timer for Cache work to do incase of Satellite Server in milliseconds
CACHETIMER=300000

You could try changing it as my conf and see if it helps.. I think it may be the CLEARTEXT=FALSE value that mig be causing this issue..

Thomas


Thanks for your input Thomas.

I will try to comment out the CLEARTEXT option. Do you think I should change the option on all our other primaries as well (although they have never had this problem)?

I am not quite sure its a good idea to change the other options in novell-zenimgweb.conf. Our four primaries all have two nics and two ipaddresses - one dedicated for client/server communication (the one that is currently configured in novell-zenimgweb.conf) The other nic/ipaddress is dedicated to DB communication.
0 Likes
Highlighted
Knowledge Partner
Knowledge Partner

Re: novell-pbserv status dead

martinusen;2064320 wrote:
Thanks for your input Thomas.

I will try to comment out the CLEARTEXT option. Do you think I should change the option on all our other primaries as well (although they have never had this problem)?

I am not quite sure its a good idea to change the other options in novell-zenimgweb.conf. Our four primaries all have two nics and two ipaddresses - one dedicated for client/server communication (the one that is currently configured in novell-zenimgweb.conf) The other nic/ipaddress is dedicated to DB communication.


Yeah if you have several nics for different purposes then leave those values intact, I've read somewhere that the cleartext=false value can cause pbserv to crash if it's under load. So if pbserv doesn't crash on the other servers then no need to change anything on them.

Thomas
0 Likes
Absent Member.
Absent Member.

Re: novell-pbserv status dead

Good morning all-

I am having the same problem with novell-pbserv. My CLEARTEXT option is commented out but I try uncommenting the ports even though I am using the standard port. It usually dies around 10-14 days.

Running 10.3.0.

# This is used by ZENWorks Imaging library that is loaded
# by ZENWorks Imaging Service
# The following are used by zeninmgweb library to connect to HTTP server

# Specify the HTTP server IPAddress
#IPADDRESS=X.X.X.X

# HTTP port
HTTP_PORT=80

# HTTPS port
HTTPS_PORT=443

# Allow HTTP if HTTPS is not available
#CLEARTEXT=FALSE

# HTTP Service Path
#HTTP_PATH=/zenworks-preboot/

# Timer for Cache work to do incase of Satellite Server in milliseconds
CACHETIMER=300000

novell-pbservd[16019]: segfault at 0000000074726543 rip 00000000f74d4b6a rsp 00000000ffc9c8c4 error 4
novell-pbservd[380]: segfault at 0000000011313544 rip 00000000f7460fd8 rsp 00000000f6bb03a8 error 4
novell-zmgprebo[28697]: segfault at 00000000164cd38c rip 000000000805297f rsp 00000000f64cd320 error 6
novell-pbservd[13901]: segfault at 000000009d650ef2 rip 00000000f74a7fd8 rsp 00000000f3ffbfa8 error 4
novell-pbservd[23868]: segfault at 00000000005827e8 rip 00000000f7485b6a rsp 00000000f3ffcf54 error 4
novell-pbservd[6974]: segfault at 0000000011313544 rip 00000000f7437fd8 rsp 00000000f1b7d5b8 error 4

JB

0 Likes
Highlighted
Knowledge Partner
Knowledge Partner

Re: novell-pbserv status dead

Okay since this also happens with cleartext commented then try setting FORCEHTTP=forced in /etc/opt/novell/zenworks/preboot/zenimgweb.conf and restart preboot services. If that doesn't work then next would be to update to 10.3.1 because I see there are a couple of bugs fixed for this issue in that release.

Thomas
0 Likes
Highlighted
Absent Member.
Absent Member.

Re: novell-pbserv status dead

We've had the same thing happen in our environment since 10.3.0. PBSERV service on some of our satellites has to be restarted at least once a week. I've had several SRs open on the issue, but no fix as of yet. For the time being we've created a cron job that checks every minute if the preboot services are dead. If they are, the services are restarted.

I see the exact errors in our logs as well.

DEV - ZCM 10.3.4/MS-SQL Database (2 Primaries & 2 Sats running 64-bit SLES 10.2 w/ ENGL Imaging Toolkit 7) PROD - ZCM 10.3.4/MS-SQL Database (5 Primaries & 11 Sats running 64-bit SLES 10.2 w/ ENGL Imaging Toolkit 7 ~7500 Managed Devices)
0 Likes
Highlighted
Absent Member.
Absent Member.

Re: novell-pbserv status dead

patrick_reeder;2066431 wrote:
We've had the same thing happen in our environment since 10.3.0. PBSERV service on some of our satellites has to be restarted at least once a week. I've had several SRs open on the issue, but no fix as of yet. For the time being we've created a cron job that checks every minute if the preboot services are dead. If they are, the services are restarted.

I see the exact errors in our logs as well.


Wed Jan 19 15:14:17 2011 -- [LicenseCheck] Checking Imaging License
Wed Jan 19 15:14:17 2011 -- [ZENIMGWEB-CSCKT] SSL_get_verify_result failed: self signed certificate in certificate chain (19)Wed Jan 19 15:14:17 2011 -- [ZENIMGWEB-CSCKT] Failed to connect to ZCM server 10.192.0.131 at port 443. Error code: 12. Certificate doesn't verify.
Wed Jan 19 15:14:17 2011 -- [ZIMGWEB-OHC] cannot connect to server : 0
Wed Jan 19 15:14:17 2011 -- [LicenseCheck] Sending OOB error message
Wed Jan 19 15:14:17 2011 -- [SCEM] OOB error message, 1855
Wed Jan 19 15:14:17 2011 -- [LC] close connection, 10.5.31.160
Wed Jan 19 15:14:17 2011 -- [ECC] CS

DEV - ZCM 10.3.4/MS-SQL Database (2 Primaries & 2 Sats running 64-bit SLES 10.2 w/ ENGL Imaging Toolkit 7) PROD - ZCM 10.3.4/MS-SQL Database (5 Primaries & 11 Sats running 64-bit SLES 10.2 w/ ENGL Imaging Toolkit 7 ~7500 Managed Devices)
0 Likes
The opinions expressed above are the personal opinions of the authors, not of Micro Focus. By using this site, you accept the Terms of Use and Rules of Participation. Certain versions of content ("Material") accessible here may contain branding from Hewlett-Packard Company (now HP Inc.) and Hewlett Packard Enterprise Company. As of September 1, 2017, the Material is now offered by Micro Focus, a separately owned and operated company. Any reference to the HP and Hewlett Packard Enterprise/HPE marks is historical in nature, and the HP and Hewlett Packard Enterprise/HPE marks are the property of their respective owners.