Dr. András Korn
Dr. András Korn
> Understood. I will add an option like `--no-split` that will tell iostat to not break a report into sub-reports so you get all the metrics on one line for...
I would submit that something like `--single-line` or `--compact` would be more expressive than `--no-split`, but it's up to you.
> > (Normally, the output is colorized so that the colons used to separate the columns from each other are barely visible -- I can't reproduce that here.) > https://asciinema.org/...
Actually, I can reproduce the problem (or a very similar one) with sync enabled. If I accidentally hide the sidebar by clicking the TST icon, then show the sidebar again,...
FWIW, a patch for coreutils that drops the LCM logic completely and limits buffer size to 32M has already been submitted, but it's probably still best to also do something...
@ryao yes, sorry: it's in https://debbugs.gnu.org/cgi/bugreport.cgi?bug=59382, the specific patch is https://debbugs.gnu.org/cgi/bugreport.cgi?att=1;filename=copy-buffer-limit.patch;bug=59382;msg=8. Update: meanwhile another, alternative fix was proposed for `cp` in the same thread.
FWIW, coreutils applied their proposed in early January 2023, so that recent versions aren't affected by this problem anymore; but it may still make sense to do something about it...
Note: `zfs set mountpoint=` has the same problem. You can't use it to explicitly set the mountpoint on a busy filesystem to what its current inherited value is unless you...
Yes, sorry, I wasn't too clear. I was thinking the IP_TRANSPARENT option could be a sort of last resort fallback when the client doesn't send SNI. So if there is...
FWIW, I also have this problem and will likely also just delete the offending entries. Just adding this comment to confirm the bug. Preferably, 1.x shouldn't export invalid entries, but...