nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

OD failure

Lal

Hi,

Recently we have added new nodes to the target servers.
After this, we are facing issues with OD.

OD server is going down quite often, is there any settings change to be done to resolve this issue,

Note: This is often noticed during peak business hours, when the number of deployments are high.

My OD version :

Interwoven OpenDeploy(6.2.1.0.0 Build 71309 Interwoven 20081020)
Component Version (6.2.1.0.0 Build 71309 Interwoven 20081020)

Could some one help me with your thoughts

Warm Regards
LAL

Find more posts tagged with

Comments

nipper

You give no information on why or what OS you are using.

Look through the logs (both OD and system, on the sender and receiver), see if there is any useful information.

You will probably need to open a ticket with support as well.

Lal

My OS:
Linux itseelm-lx40xx 2.6.9-55.ELsmp #1 SMP Fri Apr 20 17:03:35 EDT 2007 i686 i686 i386 GNU/Linux

Adam Stoller

My OS:
Linux itseelm-lx40xx 2.6.9-55.ELsmp #1 SMP Fri Apr 20 17:03:35 EDT 2007 i686 i686 i386 GNU/Linux

How much RAM, CPU, disk space are on the server(s) that appear to stop responding?
What else is running on those servers?

nipper

Who's Linux is this ? Is it a supported version (RedHat 4.0 or 5.0) ?

IWOV does not support all flavors of Linux because they can all be very different.

Lal

Hi Nipper,

Operating System: Red Hat Enterprise Linux ES release 4
(Nahant Update 5)
Installed Products: TeamSite 6.7.1 SP1 Build 68040

RAM: 7.92 GB
Hard Drive: 2*73GB (Internal)
128 GB (SAN)

// br
Lal

nipper

OK, that seems to be a solid server, what about the receivers ? Have you called support ? OD has been a workhourse and is used in very high volume applications, so this is not common.

Lal

Actually we added new target nodes recently, after that only this issue seems occurring and that too during peak business hours,

No, we are yet to contact support on this,

LAL

toggleChartSeries.rptdesign

nipper

Well see if it is specific to a certain node. Can you turn one off at a time ?

grpinterval.rptdesign

Lal

No, it wont work here, as these are production legs, there will be sync issues

//br
Lal

paul_iwov

Hi,

Recently we have added new nodes to the target servers.
After this, we are facing issues with OD.

OD server is going down quite often, is there any settings change to be done to resolve this issue,

Note: This is often noticed during peak business hours, when the number of deployments are high.

My OD version :

Interwoven OpenDeploy(6.2.1.0.0 Build 71309 Interwoven 20081020)
Component Version (6.2.1.0.0 Build 71309 Interwoven 20081020)

Could some one help me with your thoughts

Warm Regards
LAL

"OD server is going down quite often" did you notice if any hot spot logs is getting created, ti should be something hs_pid_****.log, this log would have the exact reason.

Also did you change java heap size prior to server going down ?

Lal

Hi Paul,

Where will this file located ?hs_pid_****.log , is it in the target nodes ?

How to change the java heap size ? is it in the source OD server,

Warm Regards
Lal

Firenze

Can you reproduce the error with simulated deployments?

Can you turn off transactional deployments to isolate the leg?

Are you running any DNRs?

Also, in your response to nipper about not being able to turn off a node because of production sync issues. That seems to be an architectural/design issue. Servers may go down for hardware reasons, or other, and your system should be able to recover typically by deploying an edition or similar area deployment.

Lal

How much RAM, CPU, disk space are on the server(s) that appear to stop responding?
What else is running on those servers?

Hi Fish,
Here is wat i got from out server team, sorry for the delay in reply

##
All servers have dynamic CPU allocation. That means that there is an assigned/guaranteed
number of CPUs which it will have, But if they are not in use, a server which needs more,
can borrow CPU power from another sserver. This is reflected in
the "Assigned Processor Units" and the "Assigned Virtual Processor" values.

The server is guaranteed the APU, but can use up to the "AVP" value if no other server is using them.

This is not a Windows server, so the term "disk space" is not viable. AIX uses Logical Volume Management (LVM).
Disks are assigned to a volume-group (VG) which contains disks. From the VGs,
filesystems are created of a disired size. So free "disk space" depends on what you wnat to do.

APU AVP Mem appsvg
yyseelm-axXXXX 3 5 8Gb 50Gb free
yyseelm-axXXXX 3 5 8Gb 50Gb free
yyseelm-axXXXX 2 4 8Gb 11Gb free
yyseelm-axXXXX 2 4 8Gb 11Gb free

##

Warm Regards
Lal

Adam Stoller

Hi Fish,
Here is wat i got from out server team, sorry for the delay in reply

##[indent]All servers have dynamic CPU allocation. That means that there is an assigned/guaranteed number of CPUs which it will have, But if they are not in use, a server which needs more, can borrow CPU power from another sserver. This is reflected in the "Assigned Processor Units" and the "Assigned Virtual Processor" values.

The server is guaranteed the APU, but can use up to the "AVP" value if no other server is using them.

This is not a Windows server, so the term "disk space" is not viable. AIX uses Logical Volume Management (LVM). Disks are assigned to a volume-group (VG) which contains disks. From the VGs, filesystems are created of a disired size. So free "disk space" depends on what you wnat to do.
                          APU  AVP  Mem    appsvg
yyseelm-axXXXX     3        5     8Gb    50Gb free
yyseelm-axXXXX     3        5     8Gb    50Gb free
yyseelm-axXXXX     2        4     8Gb    11Gb free
yyseelm-axXXXX     2        4     8Gb    11Gb free
[/indent]##

Warm Regards
Lal

From what I recall about AIX (it's been close to 20 years since I used it) - it still has a notion of CPU, RAM, and Disk Space - despite what your server team is saying (the concept of disk space pre-dates Windows).

It is unclear whether OD "knows" how to request "virtual CPU" if it needs it. However, even AIX boxes like the RS6000 (I know, it's about 20 years old, may not even exist anymore) had actual resident CPU on the server - and your server team should be able to tell you how much CPU is physically attached to the server.

What appears to be 8gb RAM, if I'm reading that correctly, should be sufficient.

All that being said, you would probably need to contact Support to get some help, since the way AIX is setup may, or may not have something to do with the problems you are experiencing - but you need to stress the fact that it is an AIX server when you contact them, because you have to hope to be able to get someone who is knowledgeable in both AIX and OD.

Lal

furhter analysis find that it is a dns issue.

When I gave the IP address instead of host name of the target nodes, the deploymnets became as fast as a rocket,

Lal