"vxcache stat" run on CVM slave shows incorrect information

Article:TECH202136  |  Created: 2013-01-28  |  Updated: 2013-01-29  |  Article URL http://www.symantec.com/docs/TECH202136
Article Type
Technical Solution


'vxcache stat' shows the size and usage information  for cache objects. In a CVM environment running this command on the CVM slave may show information that disagrees with the same command run on the CVM master.


 In this example, the cache object has been grown, however only the CVM master shows the new size. However, the command may show incorrect (and possibly even negative) information for any column:

master:~ # vxcache stat cache1

CACHE NAME                 TOTAL(Mb)    USED(Mb) (%)     AVAIL(Mb) (%)   SDCNT

cache1                         14336          15 (0)         14321 (99)      2



slave:~ #  vxcache stat cache1

CACHE NAME                 TOTAL(Mb)    USED(Mb) (%)     AVAIL(Mb) (%)   SDCNT

cache1                         10240           0 (0)         10240 (100)     0




 This issue effects all platforms.


The 'vxcache stat' option is not CVM aware. Hence, when the command is run on the slave it incorrectly shows local information. The CVM master maintains the correct information for cache object size and usage.  As such, the information it shows on the slave may not have been correctly updated with any changes since CVM was first started.


The issue is cosmetic and does not effect any snapshot function. Run the 'vxstat cache' command on the CVM master to get the correct information.

Version 6.0.1 contains a fix for the issue, where the CVM slave now correctly requests the information from the CVM master.

Article URL http://www.symantec.com/docs/TECH202136

Terms of use for this information are found in Legal Notices