You know, this question reminds me of that time I accidentally ran a relayout operation in production and took down the entire system. Good times, good times.
I don't know, the 'set nice' option seems like it could be a good compromise. It would allow the VxVM task to run, but with a lower priority than other tasks.
I'm not sure about that. The 'set iowait' option might be a better choice, as it would allow us to control the amount of time the VxVM task waits for I/O operations to complete.
I think the 'set iodelay' option of vxtask sounds like the best approach. It would allow us to throttle the VxVM task and reduce its impact on other activities.
Hmm, this seems like a tricky question. We need to find a way to limit the impact of the relayout operation on concurrent testing activities, while still accommodating the need to constrain the performance impact on those activities.
upvoted 0 times
...
Log in to Pass4Success
Sign in:
Report Comment
Is the comment made by USERNAME spam or abusive?
Commenting
In order to participate in the comments you need to be logged-in.
You can sign-up or
login
Fannie
9 months agoReita
9 months agoOren
9 months agoTracie
9 months agoDulce
10 months agoTracey
10 months agoTegan
10 months agoFiliberto
11 months agoMitsue
11 months agoLetha
1 years agoLorrie
1 years agoAudria
1 years agoKristal
1 years agoMurray
11 months agoChaya
11 months agoKenneth
11 months agoLinn
1 years agoDelsie
1 years ago