Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!iuvax!pur-ee!ea.ecn.purdue.edu!wieland From: wieland@ea.ecn.purdue.edu (Jeffrey J Wieland) Newsgroups: comp.sys.att Subject: Re: Re-pumping eports? Message-ID: <15982@ea.ecn.purdue.edu> Date: 28 Sep 89 17:48:24 GMT References: <1141@msa3b.UUCP> Reply-To: wieland@ea.ecn.purdue.edu.UUCP (Jeffrey J Wieland) Organization: Purdue University Engineering Computer Network Lines: 27 In article <1141@msa3b.UUCP> kevin@msa3b.UUCP (Kevin P. Kleinfelter) writes: >My 3b2/700 periodically looses one or more serial ports on an >eport card. It may lose a port or two a week on any of its eports >cards. At the present time, the only way I have found to clear >the condition is to reboot. The symptom is that no data will >move into or out of the port. I presume that I could probably >"re-pump" the eport card, but I can't seem to find the appropriate >command syntax. ... We had the same problem with the eports boards on our 3B2/500. We found that we were running release 1.1 of the eports software (which was loaded off the tape), but that we had release 1.2 on installable diskettes. When we removed 1.1 and installed 1.2, the problems pretty much went away. (We were able to use the 1.2 disks to remove the 1.1 sotware.) We still occasionally experience a hung port, probably because we running too many ports on individual boards at 9600 baud. We spread the load out across the boards; this seems to help, too. There are some fairly severe limits as to the number and speed for the ports on the ports and eports boards. You should be able to find that in the eports docs. -- Jeff Wieland wieland@ecn.purdue.edu