- CheckMates
- :
- Products
- :
- Quantum
- :
- Management
- :
- Healthcheck script issue
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
Are you a member of CheckMates?
×- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Healthcheck script issue
Hi,
I am experiencing healthcheck script (latest from 6.02 01-02-2019 as well as quite old 5.10 10-01-2018) issue, when checking cluster failovers. I can see the very same output on many clusters.
| | Number of Sync Interfaces | OK |
| | Cluster Failovers |./healthcheck.sh: line 2781: [[: 008: value too great for base (error token is "008")
./healthcheck.sh: line 2784: 008: value too great for base (error token is "008")
[Expert@FW01A:0]#
[Expert@FW01A:0]# cat /var/tmp/failovers.tmp
Nov 11 01:58:42 Master to Slave
Nov 11 01:58:42 Master to Slave
Nov 11 01:38:33 Slave to Master
Nov 11 01:38:33 Slave to Master
Nov 11 00:04:48 Master to Slave
Nov 11 00:04:48 Master to Slave
Nov 10 23:25:41 Slave to Master
Nov 10 23:25:41 Slave to Master
[Expert@FW01A:0]#
I do have suspicion, that part of the script checking, how much time since last failover, is not counting correctly, when in log is not mentioned year. In my case, failover happened on Nov 10 2018 and now is Jan 8 2019, but in log is only day and month, not year.
from healthcheck.sh:
#Determine if the year has rolled over since the failover
if [[ $current_day_of_the_year -ge $failover_day_of_the_year ]]; then
days_since_failover=$((current_day_of_the_year-failover_day_of_the_year))
else
days_since_failover=$((365-failover_day_of_the_year+current_day_of_the_year))
fi
Are you experiencing similar issue? I have attached full output from one of affected nodes.
All advises are welcome.
- Tags:
- healthcheck
- sk121447
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thank you Martin.
I have found the issue and will get an updated script added to the SK soon.
Since the failover data does not contain a year, the script calculates the number of days into the year that the failover occurred and compares it to the current day (ex: Feb 1st would be day 32). If the current day is less than the failover day, it assumes the failover occurred last year.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Updated version 6.03 01-08-2019 is working correctly.
Thank you!
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I added the related release note for 6.03 to CheckMates.
