We see these a handful of times per day, where the count is 1 or 2:
Jan 6 19:50:29 toaster-a-svif1 [tungsten-a: kern.syslog.drops:warning]: Dropped 1 syslog messages starting at Wed Jan 6 19:50:29 PST due to lack of internal log buffers
During incidents (head failovers), the issue becomes worse:
Jan 10 12:59:46 toaster-b-svif1 [toaster-b (takeover): kern.syslog.drops:warning]: Dropped 1886 syslog messages starting at Tue Jan 10 12:59:10 PST due to lack of internal log buffers
We log at severity 6, i.e. Info. Is there a way to increase internal log buffers? I use 'info' level messages to analyze problems, so I'd rather not decrease the severity threshold.
I expect the answer is 'no' ... does anyone have a work-around for this issue?
--sk
Stuart Kendrick FHCRC