05-31-2006
Quote:
Originally Posted by izy100
I dun agree to use NFS export here.
Using cluster is to achieve high availability.
Use NFS for cronjob will cause the eliminate the usefulness of clustering.
The uptime of the application will only be as good as the NFS server itself.
If using NFS is a must, use a very very high uptime Enterprise NAS is an option.
Else you can use cluster filesystem that can be mounted on both system.
PS: Failing over cronjob has been a tricky problem for Open System like Solaris and Veritas Cluster. Unless the cluster is tightly integrated into the OS, e.g. Queue Manager in OpenVMS, this problem is very very hard to solve.
Frankly speaking, I dun have a good solution too.
I would encourage the application owner to design and write their application (cronjob) in a HA manner instead of depending on the OS cron to run their job (as it is on local HDD).
I agreed with Izzy. NFS is a bad solution. THis is a common problem with an easy fix. A cron on each node must be configured. You then add a wrapper to each cron script to check which node the particular service group is running on. Veritas provides a hostactive state that you can query to determine if the service group is active. And if so, execute the script else quit.
10 More Discussions You Might Find Interesting
1. AIX
Hi,
I can use 'crontabs –e' and do all the scheduling I like. However I would like to auto send myself just the cronjobs logs that fail. That is to say the PIDs that fail and the related lines with those PID’s only. (Not the full set of logs) Has anyone done this work? Or does an AIX 5.3 tool... (0 Replies)
Discussion started by: Keith Johnson
0 Replies
2. Solaris
I run cron in solaris 10 zone. One cron job which syncing files to nfs mounted on container, creates after finishing another cron proccess(/usr/sbin/cron), and after 100 existing cron proccesses next cron job will not start. It's too weird for me, I'm not able to solve this problem. Theoretically... (3 Replies)
Discussion started by: ron76
3 Replies
3. Solaris
Hi all....I have a Sun Ultra2 that I want to use with my PC monitor. I have purchased an adapter that does not work and I was told I need to change my video card setting (if I can) to Separate Sync.....my Monitor product number ends in 1343......I am running SunOS 5.7 ......anyone have any ideas?
... (0 Replies)
Discussion started by: psantinello
0 Replies
4. Solaris
Hello All,
Anybody please help me to know ,what happens when a user having entry in both cron.allow and cron.deny files.Wheather the user will be able to access the crontab???
Thanks in advance
Vaisakh (5 Replies)
Discussion started by: ksvaisakh
5 Replies
5. Shell Programming and Scripting
Hi,
I currently use the ntpdate and date command to sync the internal clock source of my Solaris systems to the primary domain controller.
admin@myserver # ntpdate -d -u 192.168.???.???
14 Jan 17:42:02 ntpdate: ntpdate 3-5.93e+sun 03/06/05 23:16:45 (1.4)
transmit(192.168.???.???)... (2 Replies)
Discussion started by: jamba1
2 Replies
6. UNIX for Dummies Questions & Answers
Folks;
I have 2 SUSE DNS servers, one will be the master and one will be the slave. we normally update the master when adding any new servers to the DNS list.
I'm looking for a way to sync both servers every night, so the slave is updated nightly.
I thought of using "rsync" with cron to sync... (1 Reply)
Discussion started by: Katkota
1 Replies
7. Solaris
Hi ,
I have removed a cron for particular user , but cron job seems to be running even after the cron entry is removed. The purpose of the cron was to sendmail to user ( it uses mailx utility )
I have restarted cron and sendmail service still user is getting mail alerts from the cron job. And... (4 Replies)
Discussion started by: chidori
4 Replies
8. Shell Programming and Scripting
Hi
I have a requirement to write a shell script,that will check the all commented job in cron job.Please help !! (2 Replies)
Discussion started by: netdbaind
2 Replies
9. Shell Programming and Scripting
Hi,
1)
If some job supposed to run on 1st of every month at 7 AM
In cron job when we have a blackout on the 1st ( i.e when 1st falls on a sunday ) how can we make the job run the next business day?
2) How can we run a job on 25th of every quarter 7 AM(jan,apr,jul,oct)
And if 25th... (5 Replies)
Discussion started by: System Admin 77
5 Replies
10. UNIX for Dummies Questions & Answers
Hello gurus,
I am making what I think is a simple db2 call from within a shell script but I am having difficulty producing the desired
report when I run the script shown below from a shell script in cron. For example, my script and the crontab file setup
is shown below:
#!/bin/ksh
db2... (3 Replies)
Discussion started by: okonita
3 Replies
LEARN ABOUT OPENSOLARIS
scversions
scversions(1M) System Administration Commands scversions(1M)
NAME
scversions - Sun Cluster version management
SYNOPSIS
scversions [-c]
DESCRIPTION
Note -
Beginning with the Sun Cluster 3.2 release, Sun Cluster software includes an object-oriented command set. Although Sun Cluster software
still supports the original command set, Sun Cluster procedural documentation uses only the object-oriented command set. For more infor-
mation about the object-oriented command set, see the Intro(1CL) man page.
The scversions command commits the cluster to a new level of functionality after a rolling-upgrade to new Sun Cluster software. With no
arguments, the scversions command prints a message indicating whether a commitment is needed.
OPERANDS
The following operands are supported:
-c Commit the set of nodes that are currently active members of the cluster to the highest possible level of functionality.
When you upgrade a node (either through upgrade to a new release of the product or by application of a patch) and boot it back
into the cluster, some of the internal protocols on that node might have to run at lower versions in order to cooperate cor-
rectly with other nodes in the cluster. When the cluster is in this state, some administrative actions might be disabled and
some new functionality introduced in the upgrade might be unavailable.
When you run this command once from any node after all nodes are upgraded, the cluster switches to the highest versions of
internal protocols possible. Assuming that all nodes have the same Sun Cluster software installed at that time, all new func-
tionality becomes available and any administrative restrictions are removed.
If a node that has not been upgraded is an active member of the cluster at the time you run the -c option to scversions, the
command has no effect because the cluster is already running at the highest possible level of functionality.
If a node has not been upgraded and is not an active member of the cluster when you run the -c option to scversions (for exam-
ple, if that node is down for maintenance), the internal protocols of the cluster are upgraded to the highest possible ver-
sions. You might have to upgrade the node that was not an active member of the cluster to enable it to rejoin the cluster.
EXIT STATUS
0 Success
non-zero Failure
ATTRIBUTES
See attributes(5) for descriptions of the following attributes:
+-----------------------------+-----------------------------+
| ATTRIBUTE TYPE | ATTRIBUTE VALUE |
+-----------------------------+-----------------------------+
|Availability |SUNWsczu |
+-----------------------------+-----------------------------+
|Interface Stability |Evolving |
+-----------------------------+-----------------------------+
SEE ALSO
scinstall(1M)
Sun Cluster 3.2 17 Aug 2007 scversions(1M)