Status-At-A-Glance Help

Overview

Status-At-A-Glance provides the visual status of an Enstore system separated into 4 major sub-systems; the Enstore servers, the storage library, the network between the computers on which Enstore is running and any alarms that have been raised. In addition a more detailed status is given for each of the Enstore servers/movers that belong to the system. The storage library refers to one or more tape robots. Once a problem has been identified it is expected that the more complete and detailed view of the Enstore servers, the network and the alarms will be examined.

There are 4 levels of status represented on this page -

No problems indicated by a
Minor problem indicated by a
Major problem indicated by a
Situation under investigation indicated by a

In general a Minor problem will indicate that the Enstore system is still operational as a whole but some piece of it has a problem which should be investigated. A Major problem will imply that the Enstore system is not operational. Minor problems may escalate to major ones and should not be ignored. When a server (or enstore) is marked with a

it means that an unusual situation has been detected and will be monitored. Noone will be contacted until the situation has escalated to a problem.

In addition to showing the status of the system, this page will indicate when pieces of the Enstore system will be undergoing a scheduled outage by placing a after the item and including some information about the outage.

Sometimes a server or will be known to be down. In this case, the item will have a ~~line drawn through it~~.

The last section on the page is for informational purposes only. It lists the nodes in the Enstore system and the servers that run on each of them.

Enstore Overall Status

This section presents a summary of the information on the rest of the page. The following table defines what the different status levels mean for each of the sub-systems. Currently the colored ball next to the storage library can only be controlled manually.

**Status Levels**
Sub-System
enstore	All elements are marked	The Inquisitor is down Some Movers are down (less than half) or marked for a Library Manager	At least one of the following is down : Alarm Server, Configuration Server, File Clerk, Logger, Volume Clerk, a Library Manager More than one half of all Movers (for a Library Manager) are down	At least one element is marked
alarms	No alarms have been raised	One or more alarms have been raised

A next to one of these sub-systems, indicates the entire sub-system will be unavailable.

If enstore is ~~crossed out~~, the entire sub-system (the servers/movers) can be considered to be down.

Enstore Individual Server Status

This section presents the status of each individual server in Enstore. Individual servers are checked to see if they are alive. A server is not reported as being dead until it has been found to be not alive for a configurable number of times (this number is set in the configuration file under the key system and allowed_down). If a server does not seem to be alive, it is marked with a

until the number of times it is seen dead is greater than the number specified in the configuration file as mentioned above. In the following table, n represents this number.

So, for example, if there is an element in the configuration file -

configdict['system'] = { 'allowed_down' : {'log_server' : [1,10], 'library_manager' : [2,20], 'default' : [1,20] }}

then the log_server will be marked as a major problem () the first time it appears to be down. However, the following happens with respect to any library_manager -

Num times seen down in a row marker used

1

2

In the following table n refers to the number set in the configuration file (in the above example, n = 2 for the library_manager).

Status Levels
Sub-System Element

Alarm Server alive Not seen alive when checked n or more times Not seen alive when checked less than n times

Configuration Server alive Not seen alive when checked n or more times Not seen alive when checked less than n times

File Clerk alive Not seen alive when checked n or more times Not seen alive when checked less than n times

Logger alive Not seen alive when checked n or more times Not seen alive when checked less than n times

Inquisitor alive Not seen alive when checked n or more times Not seen alive when checked less than n times

Volume clerk alive Not seen alive when checked n or more times Not seen alive when checked less than n times

All Library Managers alive

More than half of the associated Movers are marked
In a pause state
In a locked state
In an ignore state.
Not seen alive when checked n or more times Not seen alive when checked less than n times

All Media Changers alive Not seen alive when checked n or more times Not seen alive when checked less than n times

All Movers alive

In a DRAINING state
In an OFFLINE state

Not seen alive when checked n or more times
In an ERROR state
Not seen alive when checked less than n times

Num times seen down in a row	marker used
1
2

**Status Levels**
Sub-System Element
Alarm Server	alive		Not seen alive when checked n or more times	Not seen alive when checked less than n times
Configuration Server	alive		Not seen alive when checked n or more times	Not seen alive when checked less than n times
File Clerk	alive		Not seen alive when checked n or more times	Not seen alive when checked less than n times
Logger	alive		Not seen alive when checked n or more times	Not seen alive when checked less than n times
Inquisitor	alive		Not seen alive when checked n or more times	Not seen alive when checked less than n times
Volume clerk	alive		Not seen alive when checked n or more times	Not seen alive when checked less than n times
All Library Managers	alive	More than half of the associated Movers are marked In a pause state In a locked state In an ignore state.	Not seen alive when checked n or more times	Not seen alive when checked less than n times
All Media Changers	alive		Not seen alive when checked n or more times	Not seen alive when checked less than n times
All Movers	alive	In a DRAINING state In an OFFLINE state	Not seen alive when checked n or more times In an ERROR state	Not seen alive when checked less than n times

Legal Notices

Last modified: Thu Mar 21 11:10:14 CST 2002