Hmm... I'll take into consideration the Mylex thing - it is half the bus speed, a third the CPU speed, and an eighth the cache. Hmm. ECC will still work without the intervention of the OS, I assume. It would be nice to know if the memory was dying, tho. I know, I know. A 486/66 can flood a line with static HTML pages. But can it handle a good number of connections, all using PHP4, making around 3 SQL queries/page, provide credit card connections, and have enough oomph for SSL also? :P Nick chrome at real-time.com, on 09/19/2000 04:19:05 PM To: tclug-list at mn-linux.org @ PMDF cc: Subject: Re: [TCLUG:21429] Server recommendations > SCSI Card: Adaptec 2100S RAID - Single channel, 32MB cache, 32-bit PCI, > Ultra160 > Will either OS let me know if a SCSI drive dies and the mirror is broken? the RAID controller we use around here is the Mylex 150 (same thing VA puts in their boxen). it reports information through a /proc interface, and I understand it will give you pretty much all the information you want, that way. if you want to monitor it with Netsaint or the like; it's certainly possible. no other controller that we know of is supported this well in the Linux kernel. (please, prove us wrong!) > Will either OS report ECC errors? I know there is a Linux project to support ECC memory; but that's all I know. as for server power; at my last job I used a 486/66 as a webserver (just static pages); and it was more than adequate. (10b-T LAN, 56K net connection). just slapping bits on the wire takes almost no CPU. :) Carl Soderstrom -- Network Engineer Real-Time Enterprises (952) 943-8700 --------------------------------------------------------------------- To unsubscribe, e-mail: tclug-list-unsubscribe at mn-linux.org For additional commands, e-mail: tclug-list-help at mn-linux.org -------------- next part -------------- --------------------------------------------------------------------- To unsubscribe, e-mail: tclug-list-unsubscribe at mn-linux.org For additional commands, e-mail: tclug-list-help at mn-linux.org