I may be wrong, but I am sure someone on this list found that the journal didn't make it back to disk with caching was on. It may have been that VM crashed now that I think about it.
-----Original Message----- From: Linux on 390 Port [mailto:[EMAIL PROTECTED] Behalf Of Ivan Warren Sent: Thursday, July 28, 2005 9:19 AM To: [email protected] Subject: Re: A file corruption causing system hang (Linux guest) > > It depends. If everything is working right, and you shut > down the system in an orderly manor then you should not have > a corruption from minidisk caching. It does interfere with > journaling, which could result in corruption if the system > crashes, or if you force the machine to logout from VM (my > terminology may be wrong). > I don't concur... The (one of the) purpose of a journaling filesystem is to prevent filesystem corruption when the operating system responsible for maintaining the said filesystem is stopped abruptly. All superblock modifications are logged into the journal, ensuring this log can be replayed, commiting or rolling back any filesystem structure modifications - thus - maintaining a clean bill of healt in the filesystem. If you logoff the virtual machine or shutdown z/VM - I very much doubt CP will just throw any uncommited minidisk write cache data out the door.. Eventually, you may have an issue if you forcefully power down the underlying hardware (or deactive the LPAR).. but this is quite extreme.. Just my €0.02 --Ivan ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
