Mantis - Resin
Viewing Issue Advanced Details
ID: Category: Severity: Reproducibility: Date Submitted: Last Update:
2099 text always 10-20-07 10:28 10-24-07 10:14
Reporter: stbu Platform:  
Assigned To: ferg OS:  
Priority: normal OS Version:  
Status: closed Product Version: 3.1.3  
Product Build: Resolution: fixed  
Projection: none      
ETA: none Fixed in Version: 3.1.4  
Summary: 0002099: resin-admin (php) might show negative uptime for one server in the cluster
Description: Resin-Professional 3.1.3 in a Load-Balancing configuration

web-a Windows XP, Java 6, 192.168.1.100
app-a Windows XP, Java 6, 192.168.1.100
app-b Linux, Java 6, 192.168.1.141

see also attached resin.conf of all three involved resin instances.


web-a and app-b are both started and I'm logged into the php resin-administration. The page cluster.php is opened and shows that app-a is not available.
I've started app-a, waited until the console displayed "Resin started in 1657ms", and pressed "refreshed".

The Uptime column of app-a showed a negative uptime as you can see from the attached screenshot and the HTML of the cluster.php page at that time.


The clock of 192.168.1.141 (app-b) and 192.168.1.100 (app-a) differ by 86 seconds.
--> This should explain the negative uptime.

BUT, it might be worth about thinking that uptime formatting in cluster.php calculate the uptime of a server using the server's startup-time *and that server's current time* rather than the time of the server of the admin page.
Steps To Reproduce:
Additional Information: Attached ZIP contains:
 - php-admin_cluster__negative_uptime_format.jpg (Screenshot with negative uptime)
 - php-admin_cluster__negative_uptime_format.html (HTML output of cluster.php)
 - resin.conf (used resin.conf)
Relationships
Attached Files:  C:\\tmp\\caucho\\negative_uptime.zip [^] (169,203 bytes) 10-20-07 10:28

There are no notes attached to this issue.