This note is a reminder that the PRocedures suggested in RFC# 381 for scheduling IMP software maintenance have now been in effect for a period of about two months. Two points apparently require re-emphasis:
1) We reserve the hour from 7 to 8 a.m. eastern time every Tuesday as the time when IMPs can be reloaded. We will probably not use this period every Tuesday, but we do reserve this period every Tuesday. The software maintenance time period is in addition to the hardware preventive maintenance scheduled at each site.
2) We now send an IMP - Going - Down message to all the Hosts at the site which contains information on the time that the IMP is going down, the eXPected duration, and the reason. We try to send out this message an hour before soft- ware releases, and five to ten minutes before scheduled pre- ventive maintenance. Further, we will also notify the Hosts at a site, whenever possible, when it is necessary to stop the IMP on an unscheduled basis. Finally, 30 seconds before the
[Page 1]
IMP goes down, a second IMP - Going - Down message will be sent to the Hosts informing them that the IMP is going down immediately. The format of the IMP - Going - Down message is as follows:
bits 1 - 4 - all zero
bits 5 - 8 - 0010
bits 9 - 16 - all zero
bits 17 - 32 - coded as follows:
All bits zero - going down in 30 seconds
Bits 17, 18 = 01 - scheduled hardware PM
Bits 17, 18 = 10 - scheduled software reload
Bits 17, 18 = 11 - emergency reload on restart
Bits 19 - 22 - how soon the IMP is going down, in 5 minute units.
Bits 23 - 32 - how long the IMP will be down, in 5 minute units.
[ This RFCwas put into machine readable form for entry ] [ into the online RFCarchives by BBN Corp. under the ] [ direction of Alex McKenzie. 1/97 ]
