This post is more than 5 years old
15 Posts
0
1908
ProSphere 1.7 deployment
Hi,
I have deployed ProSphere 1.7 using the files provided in this site:
https://community.emc.com/docs/DOC-15851
The vApp powers on properly but when logging into the web GUI two errors pop out:
1. One or more of the ProSphere virtual machines was not ready to run the ProSphere Console. Loading the Console may result in a incomplete display of data. Details: The status check for the Historical Database virtual machine has either timed out or run into errors. Check if the virtual machine´s Service Monitor is running.
2. Error ID: CAP-3001. An unknown UI Service error occurred while executing the request. HTTP fault:500
For the second error, I have read here that this can be "resolved on updating to ProSphere 1.6." but I installed version 1.7...
And also if I access to the service monitor, i.e. https:// /cgi-bin/mon.cgi everything is green.
Has anyone of you had these problems before?
ConorLawlor
233 Posts
0
February 5th, 2013 08:00
Hi,
The srm-vapp-information is the file where the appliance hostnames are found by the system when completing the status checks.
Login to the SRM Appliance host and change directory (cd) to the following location: /usr/share/tomcat6/webapps/login-monitor/WEB-INF/classes/ Run the cat command to view the contents of the srm-vapp-information file:
cat srm-vapp-information | more
Do you see the correct FQDN for HistoricalDatabase.hostname= (all lower case) and IP for HistoricalDatabase.ip=?
Regards
Conor
ConorLawlor
233 Posts
0
February 5th, 2013 03:00
Hi,
The above conditions can be found to occur in environments where DNS entries for ProSphere appliances are incorrect.
From vSphere client ensure the networking information such as DNS Servers and Search Domain for all ProSphere appliances VM is correct.
Review the following post https://community.emc.com/message/708508#708508
If the errors only appeared after a few days of a good deployment please log into the Historical Database and run the command Service --status-all and review
Regards
Conor
ejolaus
15 Posts
0
February 5th, 2013 04:00
Thanks for your answer.
I checked that post and I have the DNS properly configured in both directions (using the nsklookup it returns both the name and the ip given the ip and name respectively).
Also, the hostsnames do not contain underscores, only dashes (i.e. pro-sphere-app.xxxxx, pro-sphere-de.xxxxx, pro-sphere-hdb.xxxxx)
Also I have all those parameters that you say properly set up.
I have powered off and on again the vApp and those errors still pop out.
The errors have not appeared after a few days, it is the first time I run the GUI and would like to configure the discovers.
ejolaus
15 Posts
0
February 5th, 2013 04:00
Hi again,
I have logged into the Historical Database VM and have run service --status-all.
I found out something that maybe is a relevant:
gpstate:pro-sphere-hdb:gpadmin-[CRITICAL]:-gpstate failed. (Reason='could not connect to the server: Connection refused
Is the server running on host "127.0.0.1" and accepting TCP/IP connections on port 5432?
And then I use the command netstat to see the current ports status and 5432 does not appear. Is that normal?
ejolaus
15 Posts
0
February 5th, 2013 05:00
Yes, it seems that it has enough space:
/dev/mapper/srm-data 209596424 1640180 207956244 1% /data
Yes, the service greenplum is 'dead' and it fails when I try to start it..
ConorLawlor
233 Posts
0
February 5th, 2013 05:00
Hi
The above output is from the service greenplum status command
Seems to be an issue with the Greenplum database
Please run the df -a and confirm does the srm-data filesystem have free space available?
This issue will need support to look at.
Regards
Conor
ejolaus
15 Posts
0
February 5th, 2013 06:00
I solved the second error. It does not appear now.
The greenplum service now is working.
But now the error "The status check for the Historical Database virtual machine has either timed out or run into errors. Check if the virtual machine´s Service Monitor is running" is still there...
ejolaus
15 Posts
0
February 5th, 2013 09:00
I have redeployed the vApp again because there was a problem with the naming. I have checked this file on the StorageResourceManager and yes, the names and the IPs now are correct.
But a new problem has just come up... . When I loggin into one of the VM and have a look at the routing tables, they are empty! I remember that I did not do any network configuration for each VM.
ejolaus
15 Posts
0
February 6th, 2013 04:00
Problem solved.
There was a problem with the VM network configuration.
Thanks a lot for your help!