vCenter (v5.5.0.2001466) Service Status alert - "
Database storage is running low. Refer to VMware KB2078305 for details"
I've looked at KB2078305 -- but it is generic SQL/DB troubleshooting tips.
The vCenter Service Status alert just lists the partitons & sizes on this shared back-end SQL cluster server.
- How do i find out which drive vCenter is complaining about? The quorum/page disks will always be mostly full.
- What logfile will contain this error message on the vCenter server?
- Is there a way to configure vCenter Service Status to ignore non-DB partitions on the back-end SQL server?
- What is the low-end threshold for this alert?
I've got my DBA's looking into the SQL server - but say that all looks OK from their end. There are no failing jobs, tasks, maintenance plans.
There are no hits out there for this issue on the search engines -- so I'm turning to the vCenter Forums to see if anyone has any 411. thanks..
Peter
Hi bergerpjr77
This will partly be a "Me, too!" reply to your question, but I'll try to add some information that I found, while researching this "Database storage space is running low. Refer to VMware KB 2078305 for details." alert that started appearing in the "vCenter Service Status" page of the "Administration" area in the vSphere Client.
In the case that I'll mention, that alert started appearing after a "VMware vCenter Server" upgrade from version 5.5 ("Update 0") to 5.5 Update 2. The error message that appeared - "Database storage space is critically low and affects vCenter Server functionality. Refer to VMware KB 2078305 for details." - was slightly different from yours ("Database storage space is running low. Refer to VMware KB 2078305 for details.") and, in this case, it was an "Alert" (which, as you may know, is more severe than a "Warning" like the one you got):
I've read the "vCenter Server 5.5 Update 2 Release Notes" web page. In the "Resolved Issues -> Upgrade and Install" subsection of that page, I found the following reference:
- vCenter Server might stop functioning if the disk space on the database server is exhausted
By default, vCenter Server does not apply a retention policy on the data stored in the vCenter Server database. Over a period of time, the growth of the database files consumes all the available disk capacity and prevents the database server and vCenter Server from functioning properly.This release resolves the issue by displaying a warning in the vSphere Web Client when the database server's available disk space is low. See VMware Knowledge base article 2078305 for details.
Then, in the "Resolved Issues -> Virtual Machine Management" subsection of that same "Release Notes" page, I found the following additional information:
- vCenter Database parameters for Warning and Alert not displayed in the Advanced Settings tab*
After you perform a minor upgrade to vCenter Server 5.5 Update 2 and log in to vCenter Server using the vSphere Client or the vSphere Web Client, the Warning parameterconfig.vpxd.vdb.space.warningMB and the Alert parameterconfig.vpxd.vdb.space.alertMB are not displayed under vCenter Server Settings >Advanced Settings tab or in the vpxd.cfg file. Although, the parameters are missing, they are triggered at the default values of 5000 MB and 1000 MB respectively.Workaround: To resolve this issue, follow these steps:
- Log in to vCenter Server using the vSphere Client or the vSphere Web Client
- Go to vCenter Server Settings > Advanced Settings
- Add the parameters config.vpxd.vdb.space.warningMB for warning andconfig.vpxd.vdb.space.alertMB for alert as shown in the example here:
Key=config.vpxd.vdb.space.warningMB value=<50000> (example value in MB)
Key=config.vpxd.vdb.space.alertMB value= <2000> (example value in MB)- Click Add
So, the above paragraph answers your last question:
- Question: What is the low-end threshold for this alert?
Answer: 5000 MB for the "Warning" level and 1000 MB for the "Alert" ("critical") level.
Like in your case, this vCenter Server installation has the Database in a separate shared back-end Microsoft SQL Server cluster server, with a relatively small "Quorum" disk (a bit less than 1 GB). That apparently triggered the alert. I tried adding the "config.vpxd.vdb.space.alertMB" key in the "Advanced Settings" section of "vCenter Server Settings" using the value of 900 (MB); and - sure enough - after a few minutes the "Alert" got changed to a "Warning" (lower severity than Alert) in the "vCenter Service Status".
Now, concerning your 2nd question:
- Question: What logfile will contain this error message on the vCenter server?
Answer: After reading the VMware KB: Location of vCenter Server log files (VMware KB - Knowledge Base - article 1021804), I opened the most recent “vpxd-aaaa.log” file (where “a” is a digit) in the directory/folder "C:\ProgramData\VMware\VMware VirtualCenter\Logs" of the vCenter Server machine. The log entry that looked the most relevant to me was the following:
2014-10-15T08:49:44.616+01:00 [07572 info 'utilvpxdVdb'] [VpxdVdb::GetDBSpaceStatus] WarningThreshold(MB): 5000 AlertThreshold(MB): 1000. Updating Health Message Parameter to
--> Drive: Q, Free Space:942 MB, Used Space: 0 MB
--> Drive: I, Free Space:113413 MB, Used Space: 0 MB
--> Drive: V, Free Space:986 MB, Used Space: 0 MB
--> Drive: C, Free Space:20192 MB, Used Space: 0 MB
--> Drive: D, Free Space:21817 MB, Used Space: 0 MB
--> Drive: E, Free Space:38381 MB, Used Space: 121499 MB
--> Total Used Space: 121499 MB, Minimum Free Space: 942 MB
I have no answer - and I would also like an answer - to your other two questions:
"How do i find out which drive vCenter is complaining about? The quorum/page disks will always be mostly full."
"Is there a way to configure vCenter Service Status to ignore non-DB partitions on the back-end SQL server?"
I hope this helps a bit.
Message was edited by: Ricardo Dias Marques (added Screenshot and corrected error message for the Alert)
