lrr@Princeton.EDU (Lawrence R. Rogers) (12/07/89)
When the cube reboots, where do the messages from fsck go besides to the console, which may not be visible? Are they mirrored elsewhere for later analysis? Thanks. Larry Rogers (Postmaster@Princeton.EDU) Manager, UNIX Systems Princeton University Computing and Information Technology Computing Center 87 Prospect Street, Room 201 Princeton, NJ 08544 UUCP: princeton!lrr ARPANET/CSNET: lrr@Princeton.EDU BITNET: lrr@pucc.BITNET PHONE: 609 258 6483 FAX: 609 258 3943
dennisg@kgw2.UUCP (Dennis Glatting) (12/08/89)
In article <11993@phoenix.Princeton.EDU>, lrr@Princeton.EDU (Lawrence R. Rogers) writes: > When the cube reboots, where do the messages from fsck go besides to the > console, which may not be visible? Are they mirrored elsewhere for > later analysis? Thanks. > if you hold down the two command keys and a ` you enter into a monitor if i remember correctly if you say "msg" you get the boot messages. i have the boot window displayed on power up because i like to see rc do its thing and i'm a unix freak. watching that stuff gives me a feeling of confidence. either way there is a section in the tech docs on it. -- dennisg%kgw2%tron.UUCP@umbc3.UMBC.EDU | Dennis P. Glatting I want my own NeXT, 64 MB RAM, 660 MB SCSI, NeXT Printer. ** Accepting Donations **
izumi@violet.berkeley.edu (Izumi Ohzawa) (12/11/89)
>In article <11993@phoenix.Princeton.EDU>, lrr@Princeton.EDU (Lawrence R. Rogers) writes: >> When the cube reboots, where do the messages from fsck go besides to the >> console, which may not be visible? Are they mirrored elsewhere for >> later analysis? Thanks. Boot messages to to /usr/adm/messages (anyway, I think most of them do), along with a lot of messages the rest of the system generates. Boy, there are many unnerving messages like "Window Server died ... Cannot print..", which are apparently normal. However, I don't think full fsck is performed on normal reboots (if any), because boot messages say something like "filesystem clean -- skipping fsck ...". I think there's probably a "filesystem dirty" flag in the battery backed CMOS RAM which is cleared upon orderly shutdown, and unless this bit indicates otherwise fsck is not done on subsequent reboots. Am I correct? Izumi Ohzawa, izumi@violet.berkeley.edu
mmeyer@next.com (Morris Meyer) (12/12/89)
In article <1989Dec10.223805.17331@agate.berkeley.edu> izumi@violet.berkeley.edu (Izumi Ohzawa) writes: >However, I don't think full fsck is performed on normal reboots >(if any), because boot messages say something like "filesystem clean -- >skipping fsck ...". >I think there's probably a "filesystem dirty" flag in the >battery backed CMOS RAM which is cleared upon orderly shutdown, and >unless this bit indicates otherwise fsck is not done on subsequent >reboots. > >Izumi Ohzawa, izumi@violet.berkeley.edu The filesystem dirty bit is stored in the superblock. It is set on shutdown by the UFS code after every local inode is determined to be unreferenced and sync'd on the disk. morris meyer (mmeyer@next.com) software engineer NeXT OS Group
dennisg@kgw2.UUCP (Dennis Glatting) (12/12/89)
> I think there's probably a "filesystem dirty" flag in the > > Am I correct? > i remember reading somewhere that there is a dirty flag. it might have been the monitor docs. -- dennisg%kgw2%tron.UUCP@umbc3.UMBC.EDU | Dennis P. Glatting I want my own NeXT, 64 MB RAM, 660 MB SCSI, NeXT Printer. ** Accepting Donations **
jpd00964@uxa.cso.uiuc.edu (12/15/89)
/* Written 6:25 am Dec 12, 1989 by dennisg@kgw2.UUCP in uxa.cso.uiuc.edu:comp.sys.next */ >> I think there's probably a "filesystem dirty" flag in the >> >> Am I correct? >> > >i remember reading somewhere that there is a dirty flag. it might have >been the monitor docs. There may be, but I do know for sure that there is a clean flag put on each device when it is unmounted. It is also removed when it is mounted, so you get mildly interesting results if you actually watch the mount process. Michael Rutman SoftMed jpd00964@uxa.cso.uiuc.edu