asp@puck.UUCP (Andy Puchrik) (12/07/88)
I recently got the courage to install V2.4.0. Like others I found a few surprises. One new "feature" that really bugs me is siesta mode. The system seems to hang if I XOFF (type control-S) during screen output. Nothing I do (control-U, break, control-C) can unjam it. I tried unplugging the keyboard; it didn't help. As others have reported after a few minutes, the system "unhangs" and typeout resumes for 5-8 lines and then again wedge-a-rino. Usually after a second wait it will free up. I believe the system is hung because the disk acess light doesn't blink and the Trailblazer, which had been popping RDs and SDs, goes autistic. After the "siezure" is over, it's back to normal. Another problem that may be related to this is that when the system boots, it always takes a two minute nap. Right after the double Copyright notice and before the date/time is displayed. After nap time the system comes back up. Some people have suggested that cheap (I mean less expensive) keyboards are the problem. This appears to be another "let's fix it in hardware". (I have ordered my NS16550As.) When I run V2.3.1, which is stamped Beta Release (for V2.4) I don't see the siesta mode behavior. I'm running a PC Ltd 286-8 with the standard Maxiswitch keyboard. It has what I believe to be the latest DELL firmware: system ROM Version D11 with keyboard ROM Version D00. Another poster mentioned that the V2.4 KB driver is bigger (meaning better and probably slower) and displays the scroll lock light on the keyboard. Well my V2.3.1 systems does that just fine. A fellow V2.4 user told me he didn't see the problem much. I was skeptical. He then said that after awhile he had stopped using control-S. I have to admit I had starting piping a lot of stuff through less. Unfortunately I rolled back to V2.3.1 trying to add a second disk (another horror show) and all my bad habits (XON XOFF) returned. I linked a new kernel with the supplied dfile.wini and the new kernel still had siesta mode. I might try linking in an old kb.o. Does anyone have any ideas? Has anyone else seen this problem? -- Internet: asp@puck.UUCP Andy Puchrik uucp: decvax!necntc!necis!puck!asp Moonlight Systems ARPA: puchrik@tops20.dec.com Concord, MA 01742