Re: DS5000 memory
"Maciej W. Rozycki" <[email protected]> Tue, 4 May 2010 12:44:52 +0100 (BST)
| Newsgroups | gmane.os.netbsd.ports.pmax |
|---|---|
| Message-ID | <[email protected]> |
On Mon, 3 May 2010, Aaron J. Grier wrote: > > I vaguely remember going through this years ago, I think it was solved > > by dipping the memory module connector area in 99% isopropyl and > > inserting and removing it a couple of times (assuming this is a > > 200/240/260 with the pin/socket connectors) > > I heard that the memory sockets are only rated for a few insertion > cycles? my 5000/240 has definitely had its share of single-bit errors. > wiggling the connectors seems to postpone it for a while. It was the original PMAX (DS3100) that had memory module sockets rated for the maximum of 25 insertion-removal cycles with the recommended number limited to 5. That would certainly apply to PMIN (DS2100) systems that were virtually identical and possibly 3MIN & MAXINE (DS5000/1xx & DS5000/xx) ones that used the same modules. I recall no mention of a limit for the different-style sockets of 3MAX/+ (DS5000/2x0) systems and cleaning the contacts always fixed any bit error problems for me. Although I recall I had to be determined at least once, where I had to go through several remove-clean-insert iterations before the errors eventually went away. It could have been oxidation that happened or suchlike. You can select individual modules to test BTW, to save time that is considerable with a decently populated system -- the help message for the memory test will tell you how (I don't remember off the head, sorry). Maciej