(r.e. svclog) Menno Lageman writes: > The OP wrote that script and is now trying to get in into OpenSolaris. > The bug he filed on b.o.o last week (6530523) has the 'submitter wants > to work on bug set' to yes.
Ack! So sorry for missing this connection with Peter's mail on request-sponsor. My apologies, Peter! (Also, I went to unhide the page I made for 'SMF utilities', but noticed that I can't currently reach http://www2.petervg.nl/software. Once I can, I'll unhide the page which contains the link to your software.) The original plan in this area was to not only show the service log, but actually just show the important parts (since we can calculate when the service started failing and display messages from that point on). I'd propose putting this functionality in svcs -xv. If the service was failing/failed, we'd show the logfile from the point that the failure started by recording the file offset of the point where the failure began occurring. If there's no failure, we'd show the logfile since the start of the last method executed. (This is also all part of a grand scheme to display more unified service history, but I feel that adding this component alone to svcs -xv now wouldn't compromise the vision.) liane -- Liane Praza, Solaris Kernel Development liane.praza at sun.com - http://blogs.sun.com/lianep
