I have been working on setting up VCSA 5.5 in a lab environment for the past couple of days. I have run into a number of bugs that have caused me to re-install from scratch including the IP address reference for the InventoryService not properly changing when converting from static to DHCP and the workaround of generating new certificate information not helping.
Anyway, I now have almost everything up and running. I was able to successfully deploy cert-signed CAs onto it and add it to my demo AD domain.
However, after it comes up the vSphere Profile Driven Storage service and the vCenter Storage Monitoring Service both fail to start with messages of "Unable to retrieve health data from http://localhost/sms/health.xml "
I found some articles online suggesting that this is either due to a custom vCenter port = something I am not using - or a botched upgrade or uninstall/reinstall. This does not apply to me either. Or simply the service not trying to start up frequently enough.
I can't figure out the commands for trying to manually start this service or any other suggestions.
Has anyone else seen this or does anyone have suggestions for how to proceed with resolving these two services issues?
Kinsei, here is the KB referring what your are looking for. I wish it helps:
By the way i'm having the same issue, i tried restarting the services and it still the same. In the beginning it was working and suddenly it stopped.
The error i got is that if a browse https://<vcenterIP/DNS>:443/sps/sdk it appear "503 Service Unavailable", and my sps service is up.
What is your vcenter HW configuration? Have you tried restarting your vCenter? What vcsa version?
We keep in contact.
Hope i can help you and we can find the issue.
I'm also experiencing the same problem. I import my appliance, set the hostname and IP in the import wizard then turn it on and accept the EULA, set the SSO password and join it to the AD domain. I then follow KB2057223 to replace the certificates and that's when I get the health alerts:
VMware vSphere Profile-driven Storage Service,Alert,04/12/2013 10:33,Unable to retrieve health data from http://localhost/sps/health.xml
VMware vCenter Storage Monitoring Service,Alert,04/12/2013 10:33,Unable to retrieve health data from http://localhost/sms/health.xml
Visiting the URL gives me the same "503 Service Unavailable" mentioned by fcardarelli.
Edit: I've followed this procedure twice with the same results.
Using Francisco's link I was able to start the VMWare Profile Drive Storage Service (vmware-sps) from the CLI. When checking the service via the CLI command it says that is is started.
"VMware vSphere Profile-Driven Storage Service is running (PID:14732, Wrapper:STARTED, Java:STARTED)"
However, the GUI still shows that is is down within VCSA.
Also, I don't see a service name listed for the VMWare Storage Monitoring Service in that article and cant determine the correlating service name for it.
VCSA hardware configuration is just a VM installed with the OVA. I have rebooted many many times. It has no effect.
The services appear to be started for me as well but still show down.
I've tried to log a call but the form wasn't working for me yesterday so I'll have to try again.
Before I reverted back, to my snapshot that I took before installing certificates, I had a look in the log files and found that the sps service was failing to log itself in over and over. I believe it uses certificates to log in so maybe there is a further certificate that needs changing or re-registering.
I'm having this issue, too. I ended up ditching the VCSA I had and starting a new one from scratch, only to end up with the same issues. Something I did, that in retrospect maybe I shouldn't have done, is update to the most current version of VCSA. I'm at 126.96.36.19900 Build 1891314. I'm using the self-signed certificates.
Has anyone had a resolution to this yet?
I had the problem where the SMS and SPS services were not starting. If I manually started the vmware-sps service it would start and then stop about 10 seconds later. I was getting the warnings:
Unable to retrieve health data from http://localhost/sms/health.xml
Unable to retrieve health data from http://localhost/sps/health.xml
I did VMware KB: vCenter Inventory Service fails to start in vCenter Server Appliance 5.x and everything started working again even though my Inventory Service was started and appeared to be working, but just the Storage Monitoring Service and Profile-Driven Storage service were not starting.
I'm not clear on how to update the version of OpenSSL. I've Googled for it but the stuff that came up looked like instructions for reinstalling the certs, not updating the OpenSSL version.
I resolved this issue using the following KB @ VMware. I was replacing self-signed certificates at the time and had a brand new VCSA installation.
I hope this isn't thread-jacking, please let me know if I should post this in a new thread. I'm seeing this error repeated in my ds.log file:
[2014-07-29 13:23:55,450 pool-19-thread-9 ERROR com.vmware.vim.dataservices.provider.VcProvider] Cannot login: com.vmware.vim.binding.vim.fault.InvalidLogin:
inherited from com.vmware.vim.binding.vim.fault.VimFault:
inherited from com.vmware.vim.binding.vim.fault.InvalidLogin: Cannot complete login due to an incorrect user name or password.
I've used the same root user/password for setting up SSO. I'm not clear on what login it is complaining about.
I'm not replacing any certificates or anything. All my services seems to running OK, but i'm still getting the complaints in the web client under vCenter>Monitor>Service Health and no inventory.
The complaints are as follows:
License Services>Asset Properties History services - cannot obtain asset properties from vCenter Server query service
VMware vSpere Profile-driven Storage service - Unable to retrieve heath data from http://localhost/sps/health.xml
VMware vCenter Storage Monitoring Service - unable to retrieve health data from http://localhost/sms/health.xml
Message was edited by: scottdillard - added more information about issue.
I resolved my issue. I'll try to keep it short.
I had everything up and going at one point without any issues. I monkeyed with some network settings and that's when my problems started. I tried redeploying VCSA several times but kept getting the same errors that indicated some problems with certificates, and my web client was empty of inventory with the exception of the VCSA being present.
There are two things I did differently that could be the issue. First, I had been updating the VCSA during the configuration process. This last time I did not run any updates to VCSA(I'm at version 5.5.0 build 1476327).
The second thing I did differently is that I configured my environment through the web client first. I added all my hosts and stuff through the web client before I ever connected with the Windows Client. Now all my inventory shows up and everyone is happy.
I'm not really sure which change caused it to work, but it works.