Xymon Mailing List Archive search

disk monitoring

12 messages in this thread

list Michael Magua · Mon, 21 Aug 2006 17:57:52 +0200 ·
Hi

I have specified on the server to monitor a host's disks however it is
not working. On the server I have in ~/server/etc/bb-hosts:

10.1.1.254 mail # conn ssh smtp pop3 disk

And in ~/server/etc/hobbit-clients.cfg I have:

HOST=mail
        DISK    / 90 95
        DISK    /boot 90 95

This is not being reported...

Thanks
Michael
list Michael Magua · Mon, 21 Aug 2006 17:59:24 +0200 ·
quoted from Michael Magua
On 8/21/06, Michael Magua <user-8758490797d5@xymon.invalid> wrote:
Hi

I have specified on the server to monitor a host's disks however it is
not working. On the server I have in ~/server/etc/bb-hosts:

10.1.1.254 mail # conn ssh smtp pop3 disk

And in ~/server/etc/hobbit-clients.cfg I have:

HOST=mail
        DISK    / 90 95
        DISK    /boot 90 95

This is not being reported...

Thanks
Michael
Oh and on the client I have this in the log file:

[hobbit at mail ~]$ tail -f client/logs/hobbitclient.log
2006-08-21 17:45:10 Failed to get a message, terminating
2006-08-21 17:50:11 Failed to get a message, terminating
2006-08-21 17:55:11 Failed to get a message, terminating
list Buchan Milne · Tue, 22 Aug 2006 09:58:02 +0200 ·
quoted from Michael Magua
On Monday 21 August 2006 17:59, Michael Magua wrote:
On 8/21/06, Michael Magua <user-8758490797d5@xymon.invalid> wrote:
Hi

I have specified on the server to monitor a host's disks however it is
not working. On the server I have in ~/server/etc/bb-hosts:

10.1.1.254 mail # conn ssh smtp pop3 disk

Monitoring of disks does not need to be configured on the server. If the client sends the data, it will be monitored, and you should see a disk column, with the data reported by the client.
quoted from Michael Magua
And in ~/server/etc/hobbit-clients.cfg I have:

HOST=mail
        DISK    / 90 95
        DISK    /boot 90 95
This is only the threshholds, and it won't help until you have some disk usage data reported by the client.
quoted from Michael Magua
This is not being reported...

Thanks
Michael
Oh and on the client I have this in the log file:

[hobbit at mail ~]$ tail -f client/logs/hobbitclient.log
2006-08-21 17:45:10 Failed to get a message, terminating
2006-08-21 17:50:11 Failed to get a message, terminating
2006-08-21 17:55:11 Failed to get a message, terminating
This may indicate that the client can't connect to the hobbit server. Test from the client with something like:

su - hobbit -c '$HOME/bin/bb xxx.xxx.xxx.xxx ping'

(replace xxx.xxx.xxx.xxx with the IP address or hostname of your hobbit server as you have configured it on the client).

You should get something like this back:
hobbitd 4.2.0

If you get an error, you should look at the usual culprits (if you used a hostname, test with the IP, if it doesn't work with the IP, check firewalls etc).

If it does work, it may be that the entry for the host in bb-hosts differs from the hostname the client is sending, check in the Ghost reports (Reports->Ghost Clients), and either correct the bb-hosts entry, the hostname the client uses, or map the hostname the client sends to its bb-hosts entry with the CLIENT keyword see bbhosts(5).


Regards,
Buchan

-- 
Buchan Milne
ISP Systems Specialist
B.Eng,RHCE(803004789010797),LPIC-2(LPI000074592)
list James Wade · Fri, 20 Jul 2007 11:17:03 -0500 ·
Hello All,

 
I had a recent problem with monitoring filesystems, and

I wanted some assistance.

 
We have a couple database servers using Veritas clustering.

Can anyone recommend a method for monitoring the Veritas

clustering and filesystems?

 
When the clustering went down, several partitions were unmounted.

Hobbit didn't alert us that the partitions were not available.

 
I'm curious, let's say hobbit is monitoring the filesystem (disk),

and someone unmounts a partition that is in vfstab or the filesystem

is no longer accessible, disk goes down, but the mount still shows.

Does Hobbit alert to these?

(The cluster mounts are not in the vfstab, but I just wondering

about other filesystems)

 
Thanks..James
list Bruce White · Fri, 20 Jul 2007 11:30:14 -0500 ·
James,

 
There is a script available that will check that files systems on an HPUX
that are in the /etc/fstab file are mounted.  However, this will not help
you in a cluster situation.   I have not seen a hobbit script for doing a
check of a VCS cluster.  At my last job we used BMC Patrol and we wrote
custom scripts to check the health of both our VCS and Service Guard
clusters.   I am about to start a project where I will be creating 2 Service
Guard clusters in Production.   If I don't find a hobbit/BB script to
monitor their health, I will be creating one.

 
         ......Bruce
quoted from James Wade

 
-----Original Message-----
From: James Wade [mailto:user-659655b2ea05@xymon.invalid] 
Sent: Friday, July 20, 2007 11:17 AM
To: user-ae9b8668bcde@xymon.invalid
Subject: [hobbit] Disk Monitoring

 
Hello All,

 
I had a recent problem with monitoring filesystems, and

I wanted some assistance.

 
We have a couple database servers using Veritas clustering.

Can anyone recommend a method for monitoring the Veritas

clustering and filesystems?

 
When the clustering went down, several partitions were unmounted.

Hobbit didn't alert us that the partitions were not available.

 
I'm curious, let's say hobbit is monitoring the filesystem (disk),

and someone unmounts a partition that is in vfstab or the filesystem

is no longer accessible, disk goes down, but the mount still shows.

Does Hobbit alert to these?

(The cluster mounts are not in the vfstab, but I just wondering

about other filesystems)

 
Thanks....James

 

Note: The information contained in this message may be privileged and
confidential and protected from disclosure. If the reader of this message is
not the intended recipient, or an employee or agent responsible for
delivering this message to the intended recipient, you are hereby notified
that any dissemination, distribution or copying of this communication is
strictly prohibited. If you have received this communication in error,
please notify us immediately by replying to the message and deleting it from
your computer. Thank you. Fellowes, Inc.
list Dean Casey · Fri, 20 Jul 2007 14:05:23 -0400 ·
I believe there is an "hastatus" script on DeadCat that works for
monitoring VCS clusters. IIRC, we edited that for use with VCS
clustering here. Had to put sticky bit permissions on "hastatus" & a
couple of other VCS commands to make the script work when ran by Hobbit
user), and also sticky bit on the script itself.

 
Dean Casey
quoted from Bruce White


From: White, Bruce [mailto:user-58f975e8bf9d@xymon.invalid] 
Sent: Friday, July 20, 2007 12:30 PM
To: 'user-ae9b8668bcde@xymon.invalid'
Subject: RE: [hobbit] Disk Monitoring

 
James,

 
There is a script available that will check that files systems on an
HPUX that are in the /etc/fstab file are mounted.  However, this will
not help you in a cluster situation.   I have not seen a hobbit script
for doing a check of a VCS cluster.  At my last job we used BMC Patrol
and we wrote custom scripts to check the health of both our VCS and
Service Guard clusters.   I am about to start a project where I will be
creating 2 Service Guard clusters in Production.   If I don't find a
hobbit/BB script to monitor their health, I will be creating one.

 
         ......Bruce

 
-----Original Message-----
From: James Wade [mailto:user-659655b2ea05@xymon.invalid] 
Sent: Friday, July 20, 2007 11:17 AM
To: user-ae9b8668bcde@xymon.invalid
Subject: [hobbit] Disk Monitoring

 
Hello All,

 
I had a recent problem with monitoring filesystems, and

I wanted some assistance.

 
We have a couple database servers using Veritas clustering.

Can anyone recommend a method for monitoring the Veritas

clustering and filesystems?

 
When the clustering went down, several partitions were unmounted.

Hobbit didn't alert us that the partitions were not available.

 
I'm curious, let's say hobbit is monitoring the filesystem (disk),

and someone unmounts a partition that is in vfstab or the filesystem

is no longer accessible, disk goes down, but the mount still shows.

Does Hobbit alert to these?

(The cluster mounts are not in the vfstab, but I just wondering

about other filesystems)

 
Thanks....James

 
Note: The information contained in this message may be privileged and
confidential and protected from disclosure. If the reader of this
message is not the intended recipient, or an employee or agent
responsible for delivering this message to the intended recipient, you
are hereby notified that any dissemination, distribution or copying of
this communication is strictly prohibited. If you have received this
communication in error, please notify us immediately by replying to the
message and deleting it from your computer. Thank you. Fellowes, Inc.

list Dean Casey · Sat, 21 Jul 2007 09:33:08 -0400 ·
You could also set hobbit as an administrative user in VCS, as an
alternative to putting sticky bit on the monitoring script.

 
Dean Casey 
quoted from Dean Casey


From: Dean Casey [mailto:user-99832ab50ffd@xymon.invalid] 
Sent: Friday, July 20, 2007 2:05 PM
To: user-ae9b8668bcde@xymon.invalid
Subject: RE: [hobbit] Disk Monitoring

 
I believe there is an "hastatus" script on DeadCat that works for
monitoring VCS clusters. IIRC, we edited that for use with VCS
clustering here. Had to put sticky bit permissions on "hastatus" & a
couple of other VCS commands to make the script work when ran by Hobbit
user), and also sticky bit on the script itself.

 
Dean Casey


From: White, Bruce [mailto:user-58f975e8bf9d@xymon.invalid] 
Sent: Friday, July 20, 2007 12:30 PM
To: 'user-ae9b8668bcde@xymon.invalid'
Subject: RE: [hobbit] Disk Monitoring

 
James,

 
There is a script available that will check that files systems on an
HPUX that are in the /etc/fstab file are mounted.  However, this will
not help you in a cluster situation.   I have not seen a hobbit script
for doing a check of a VCS cluster.  At my last job we used BMC Patrol
and we wrote custom scripts to check the health of both our VCS and
Service Guard clusters.   I am about to start a project where I will be
creating 2 Service Guard clusters in Production.   If I don't find a
hobbit/BB script to monitor their health, I will be creating one.

 
         ......Bruce

 
-----Original Message-----
From: James Wade [mailto:user-659655b2ea05@xymon.invalid] 
Sent: Friday, July 20, 2007 11:17 AM
To: user-ae9b8668bcde@xymon.invalid
Subject: [hobbit] Disk Monitoring

 
Hello All,

 
I had a recent problem with monitoring filesystems, and

I wanted some assistance.

 
We have a couple database servers using Veritas clustering.

Can anyone recommend a method for monitoring the Veritas

clustering and filesystems?

 
When the clustering went down, several partitions were unmounted.

Hobbit didn't alert us that the partitions were not available.

 
I'm curious, let's say hobbit is monitoring the filesystem (disk),

and someone unmounts a partition that is in vfstab or the filesystem

is no longer accessible, disk goes down, but the mount still shows.

Does Hobbit alert to these?

(The cluster mounts are not in the vfstab, but I just wondering

about other filesystems)

 
Thanks....James

 
Note: The information contained in this message may be privileged and
confidential and protected from disclosure. If the reader of this
message is not the intended recipient, or an employee or agent
responsible for delivering this message to the intended recipient, you
are hereby notified that any dissemination, distribution or copying of
this communication is strictly prohibited. If you have received this
communication in error, please notify us immediately by replying to the
message and deleting it from your computer. Thank you. Fellowes, Inc.

list Christian Rehberger · Mon, 17 Sep 2007 17:01:46 +0200 ·
Hello,

I'm running a hobbit server and installed a new hobbit client on a new machine.

Now I run into problems with the disk monitoring.

The disk is filled up to 93% on the new host "nestor" but on the monitoring site
I get always a yellow page. The entry for nestor in the "hobbit-clients.cfg" on the
server shows, that only the default value is working for that new client.

The client on galway works pretty fine!

On the same time I get error messages on the client machine in 
the "hobbitclient.log" file: Failed to get a message, terminating

Any idea?

Here's the "hobbit-clients.cfg" file of the server:

...
HOST=galway
        # These are the values for galway.Sisis.de
        DISK    * 97 99
        PORT    "LOCAL=%([.:]5000)$" state=LISTEN TEXT=ASE TRACK=ase
        
        PROC    /usr/local/apache2/bin/httpd 5 20 TEXT=HTTP-ContentDM
        PROC    /usr/local/sisis-pap/apache/bin/httpd 5 20 TEXT=HTTP-sisis-pap
        FILE    /home/sisis/hugo MODE=444 yellow
        LOG     /var/log/messages %hugolein color=red

HOST=nestor
        # These are the values for nestor.Sisis.de
        DISK    * 95 98

DEFAULT
        # These are the built-in defaults.
        UP      10
        LOAD    8.0 10.0
        DISK    * 90 95
        MEMPHYS 100 101
        MEMSWAP 50 80
        MEMACT  90 97

Tia and best regards

Chris

-- 
Christian Rehberger
System Consultant · OCLC PICA GmbH
Grünwalder Weg 28g · 82041 Oberhaching · Germany
t +49-(0)89-61308 333 · f +49-(0)89-61308 399 
e user-612b35aa5f4b@xymon.invalid · w http://www.oclcpica.org

OCLC PICA GmbH
Geschäftsführer: 
Christine Magin-Weeger,
Norbert Weinberger
Sitz der Gesellschaft: Oberhaching 
HRB München: 113261
list Henrik Størner · Mon, 17 Sep 2007 17:30:45 +0200 ·
Hi Christian,

the "Failed to get a message" error indicates that you're running the
client in "local mode", i.e. it uses a hobbit-clients.cfg which is maintained on the CLIENT, not on the Hobbit server. So you should
either change nestor to run in the normal (non-local) mode, or
update the hobbit-clients.cfg on nestor.

Oh, and the "Failed to get a message" error is harmless.


Regards,
Henrik
quoted from Christian Rehberger

On Mon, Sep 17, 2007 at 05:01:46PM +0200, Christian Rehberger wrote:
I'm running a hobbit server and installed a new hobbit client on a new machine.

Now I run into problems with the disk monitoring.

The disk is filled up to 93% on the new host "nestor" but on the monitoring site
I get always a yellow page. The entry for nestor in the "hobbit-clients.cfg" on the
server shows, that only the default value is working for that new client.

The client on galway works pretty fine!

On the same time I get error messages on the client machine in the "hobbitclient.log" file: Failed to get a message, terminating

Any idea?

Here's the "hobbit-clients.cfg" file of the server:

...
HOST=galway
        # These are the values for galway.Sisis.de
        DISK    * 97 99
        PORT    "LOCAL=%([.:]5000)$" state=LISTEN TEXT=ASE TRACK=ase
                PROC    /usr/local/apache2/bin/httpd 5 20 TEXT=HTTP-ContentDM
        PROC    /usr/local/sisis-pap/apache/bin/httpd 5 20 TEXT=HTTP-sisis-pap
        FILE    /home/sisis/hugo MODE=444 yellow
        LOG     /var/log/messages %hugolein color=red

HOST=nestor
        # These are the values for nestor.Sisis.de
        DISK    * 95 98

DEFAULT
        # These are the built-in defaults.
        UP      10
        LOAD    8.0 10.0
        DISK    * 90 95
        MEMPHYS 100 101
        MEMSWAP 50 80
        MEMACT  90 97

Tia and best regards

Chris

-- 
Christian Rehberger
System Consultant · OCLC PICA GmbH
Grünwalder Weg 28g · 82041 Oberhaching · Germany
t +49-(0)89-61308 333 · f +49-(0)89-61308 399 e user-612b35aa5f4b@xymon.invalid · w http://www.oclcpica.org

OCLC PICA GmbH
Geschäftsführer: Christine Magin-Weeger,
Norbert Weinberger
Sitz der Gesellschaft: Oberhaching HRB München: 113261 

-- 

Henrik Storner
list Michael C. Schultheiss · Sat, 15 Jun 2019 13:01:18 +0000 ·
We had one of our postgres database servers go offline this morning since its data partition got 100% full. Xymon didn't alert us since the xymon client was running as user xymon and didn't have permission to the postgres data directory (i.e. df gave a permission denied). I worked around it by changing the client on this server to run as root rather than xymon but is there a better way to give the client privileged access to do its checks than running the client as root?


--
Michael Schultheiss, RHCSA, RHCE   OIT Administration
Sr. Unix Systems Engineer          Ivy Tech Community College
user-4a65965dd360@xymon.invalid
list Paul Root · Sat, 15 Jun 2019 15:18:44 +0000 ·
Give a group rights to the file system and add xymon to that group.
quoted from Michael C. Schultheiss

From: Xymon <xymon-bounces at xymon.com> On Behalf Of Michael C. Schultheiss
Sent: Saturday, June 15, 2019 8:01 AM
To: xymon at xymon.com
Subject: [Xymon] Disk monitoring


We had one of our postgres database servers go offline this morning since its data partition got 100% full. Xymon didn't alert us since the xymon client was running as user xymon and didn't have permission to the postgres data directory (i.e. df gave a permission denied). I worked around it by changing the client on this server to run as root rather than xymon but is there a better way to give the client privileged access to do its checks than running the client as root?


--
Michael Schultheiss, RHCSA, RHCE   OIT Administration
Sr. Unix Systems Engineer          Ivy Tech Community College

user-4a65965dd360@xymon.invalid<mailto:user-4a65965dd360@xymon.invalid>
This communication is the property of CenturyLink and may contain confidential or privileged information. Unauthorized use of this communication is strictly prohibited and may be unlawful. If you have received this communication in error, please immediately notify the sender by reply e-mail and destroy all copies of the communication and any attachments.
list Schminke_Erik_D · Mon, 17 Jun 2019 13:46:31 +0000 ·
What I have found is that the user running "df" against a given file system must have execute permissions on every directory from the parent directory of the mount point all the way back to /

e.g.

For file system mounted at /really/long/path/mountpoint

All directories leading up to it...

/really/long/path
/really/long
/really
/

...must have +x permissions.

Whether you do that at owner, group, or other permissions is a determination you're going to have to make for yourself.  You could add the xymon user to the group assigned to those directories, but if the group has write access to those directories, that could be a security liability for your database.  You could do it on "other" permissions, but that would give everyone with access to your system execute access to those directories, giving them the ability to "cd" into them.  That may or may not be a problem for you.  The third option would be to look into whether your system supports ACLs and through those give just the xymon user +x permissions to just those directories.  Were it me doing it, that would be my preference.


Erik D. Schminke | Associate Systems Programmer
Hormel Foods Corporation | One Hormel Place | Austin, MN XXXXX
Phone: (XXX) XXX-XXXX
user-15513f33c451@xymon.invalid | www.hormelfoods.com