home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!CCM.HF.INTEL.COM!PATRICK_N_WILLIAMS
- Return-Path: <@VM1.CC.UAKRON.EDU,
- @VM1.CC.UAKRON.EDU:ccm!Patrick_N_Williams@intelhf.intel.com>
- Message-ID: <930122061102_1@ccm.hf.intel.com>
- Newsgroups: bit.listserv.banyan-l
- Date: Fri, 22 Jan 1993 06:11:02 PST
- Sender: Banyan Networks Discussion List <BANYAN-L@AKRONVM.BITNET>
- From: Patrick N Williams <Patrick_N_Williams@CCM.HF.INTEL.COM>
- Subject: Re: lost servers
- Lines: 16
-
- I've also seen this problem. I'm not sure why it happens, but the way that we
- fixed it is to do a Coldstart (using Vines Assistant, for example) on all the
- servers that were not seeing ALL other servers. In other words, your server B.
- Banyan were not sure why it happened, but when we did this, it worked fine.
- Before we knew about Coldstart, the other way we used was to put server A (the
- good one) on the same segment as server B (the bad one), and isolate that
- segment from the rest of the LAN. We then put the clocks forward at least 96hrs
- on server B. This caused the STHeardFrom table to rebuild itself. So it would
- then suck the STHeardFrom from server A. Then we just changed the time back on
- server B, and made sure the time on both servers before putting them both back
- onto the LAN. Very crude, but it worked.
- The best option, of course, is the former.
-
- Pat N Williams
- Network Chappie
- Intel Corp (UK) Ltd.
-