0% found this document useful (0 votes)
19 views

1.1 Mod7 - Logfiles PDF

The document discusses Netapp cluster log files, how they are stored in different locations like /var and /vol0, and how they can be viewed, collected, and sent to a central log server or via autosupport. It also covers mounting the root volume of a failed node using the mount_partner command to access logs and other data.
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
19 views

1.1 Mod7 - Logfiles PDF

The document discusses Netapp cluster log files, how they are stored in different locations like /var and /vol0, and how they can be viewed, collected, and sent to a central log server or via autosupport. It also covers mounting the root volume of a failed node using the mount_partner command to access logs and other data.
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Mod7_logfiles

We will have a look at where they are located and how you can collect
them to send them somewhere else. We will see how to access them via
a Browser..,Also we will look at how you can automatically have them
sent to a logserver somewhere on your network
and we will see autosupport and how logfiles are sent to Netapp.

Logfiles can be found, in two locations. On the one hand we have the
var filesystem and on the other hand we have vol0.
The var filesystem is used to log at boottime and it will store what
happens to the node, before vol0 is mounted.

After vol0 is mounted it will start logging to vol0. Remember vol0


is mounted on /mroot. The logfiles can be found in two directories
in /mroot: /mroot/etc/log and /mroot/etc/log/mlog.

To view all of the logfiles you can go to the systemshell and access
or send the files to any location. In this picture you see that the
/mroot/etc/log directory contains 250M worth of logfiles. To collect
them in one single zipped file you can use tar. c stands for create,
z stands for zip and f stands for the file to generate, and of course
the dot stands for the current directory So all files from this directory
and its subdirectory will be stored in the log_collection file.
You can send this file with for example scp to any location.

Or you can start a browser and access the logfiles of vol0 by connecting
to the Service processor infrastructure.

So let's do that....

From the clustershell you can view the log information using : event log
show, and there are some parameters you can enter like a time period for
example that will allow you to filter, or the severity of the message. This
type of logging is like in most other Unixes. Basically event log show will show
events that occurred in the cluster.

You can also run debug log show, this will show you the merged logfiles from
all nodes in the cluster.

To send your logfiles to a central loghost you can run the following command:

cluster log-forwarding modify -destination <loghost> -port <port>

Autosupport. All logfiles and more will be sent to Netapp by the Autosupport
tool. By default, autosupport will send a fill report of your cluster to netapp
every Sunday night. And of course if something break, like a port or a disk, then
that will trigger messages to Netapp as well. You can also invoke a message yourself,
for example if you want to test something and not want to alarm netapp...
The basic settings are pretty simple but an awful lot can be configured...

The last thing I want to share with you is that if a node is down, then the
partner in the Ha pair will takeover the aggregates of the failed node.
It will also take the rootaggregate of that node.
To get to the content of vol0 of that node, for example if you want to look at
configuration things or logs or coredumps or anything in the node's vol0
you can mount the volume using the mount_partner command from the systemshell
the volume will then be mounted to /partner and you can access it
Now I sincerely hope you have enjoyed this training.
If you have any remarks or questions, you can always contact me.

You might also like