From my experience if you run a metastat|grep replace
and a "metareplace device" is listed I would run that command as suggested to clear problem. But If a metastat shows that a device is in maintenance only, with no metareplace then it's no good running a metareplace.
If I do a grep for "replace" running a metastat it will not find anything because the hot spare has recovered the failed drive and everything appears to be OK.
I would do a metareplace simply because it is less work. Remember that you need to set up the slices on the replacement disk first before using it. You may also need to fix some of your metadb copies if they were using the broken disk.
I found a link that describes the steps to replacing a failed drive which seems quite good. I guess there are many ways to tackle this problem but I'm looking for the quickest and easiest. Thank you Annihilannic for the advice. Here is the a good link:
anyone ever ran into an issue when metastat stated specific metadevices were in 'need maintenance' mode, but after thorough checking and verify that the disk in fact seems to be in good working order. I have seen it a few times, but my real question is about using the metareplace to replace the same disk. From what understand it is not possible to replace it with itself, is that true?
just thought I would ask since we are on the metareplace topic.
I've seen that too with intermittent disk issues, for example a loose cable, where the disk itself is perfectly healthy. In that case you can safely replace the disk with itself using metareplace -e.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.