Dec 2011 - Guidedog Failed. 12/11 (Sunday) Eric V called at 8:30pm. Guidedog (a sparc 5/64) unable to boot. At OpenBoot sending out ARP request. Night will be weathered out - tony to addressed on Monday. 12/12 (Monday) Make a restore disk: 1212-restore.txt The guidedog disk drive is OK after doing filecheck on planck2. Also tried running guidedog with at restored disk from the 2011/09/16 backup. OBP (OpenBoot PROM) says 'IDPROM contents are invalid'. This indicate bad NVRAM or battery. Search on: IDPROM contents are invalid http://lib.ru/TXT/faqsunnvram.txt http://www.basesoft.se/ssl/solaris_howto.html Able to boot guidedog after entering configuration in OBP by hand. (see 1211-restore.txt) VME access do not work: Can't talk to dsp board or GPS board. 1212-pics - tried taking pictures of guidedog's OPB screen. 12/13 (Tuesday) Guidedog shut it self down overnight. Daycrew found the guidedog fan was not working - replaced fan. Manao shipping spex development system to hilo. (pickup about noon, should arrive tomorrow). Used some old sparc data disk to make a boot disk for hammer and hilodog: 1212-restore.txt Hilo has a USPIIi- 1v vme computer (hilodog). This system started to freeze after a lighting storm in Nov2011. Spend some time trying to get it running -> computer will not display on the monitor! 12/14 (Wednesday) TD picks boxes a FedEx (hammer VME crate, manuals, etc) and take it to the summit 1214-pics Pictures of guidedog before taking any boards out. Pictures of hammer before taking any boards out. Pictures of hammer OBP screen before taking any boards out. Realized that hammer only as SCSI-2 interface and 10BaseT Ethernet. Attached a restored guidedog disk to hammer's external scsi port (with appropriately scsi adapters). Was going to try to book off external disk, but once we tried this we got error in the OBP indicating SCSI bus error, possible blown Fuse. Took hammer VME system and the guidedog CPU down to hilo. 12/15 (Thur) Fixed fuse on hammer (by taking the one from guidedog's motherboard). Move the SCSI/Ethernet SBus card from guidedog to hammer. Booted hammer - SCSI errors when away. Able to boot up original guidedog boot disk (ST 318275LW). Was trying all afternoon to get another guidedog disk restored using all the left over sun SCSI disk Miranda saved. They can be formatted and setup on magellan (Sun Blade 100, solaris 8), but none of them work with hammer. Electrical spec sheet on the drives are the same (but the ST318275LW is an older series). (errors indicate Single-Ended vs Low Voltage Differential). Don't what to risk damage to scsi interface, just go with fixed_hammer & original guidedog disk. 12/16 (Friday) Taking up fixed_hammer_cpu_board + external disk (old guidedog disk). Installed hammer with external disk box, booted, and started up guidedog. Guidedog's software started up OK, and imaged the array OK. Called John Rayner for check out. - SPEX WORKS! In manoa, DW put a new fuse on guidedog. Starts (IPROM is bad, searching for ARP packages) Tried booting off of hammer's old disk, but it is corrupted. 1216-pics - pictures of fixed summit system. Follow UP Picture of the system. guidedog.html - description of current guidedog guidedog.dmesg.txt - output of 'dmesg' guidedog.prtconf.txt - output of 'prtconf -vp' bigdog.html - description of current bigdog bigdog.dmesg.txt - output of 'dmesg' bigdog.prtconf.txt - output of 'prtconf -vp' Dec17 dmesg from guidedog dmesg.guidedog.txt dmesg from bigdog dmesg.bigdog.txt I'm thinking miranda's disk are LVD, and spex just needs single ended. so guideguide/bigdog has ST318275LW/ST39173W. I will by 2 used ST39173W (ultrawide 9GB 68-pint HD D-Sub). Ordered from: 360 Technologies (www.360tech.com), and PartsArcade.com Dec18 Tony damaged Hammer's VME backplane (electrical short). The picture below shows damage to the J2 (bottom) connect, slot 1, Pin C1. What happen: Guidedog motherboard is a sparc 5/64. I remove it from slot 1, and replaced it with Hilodog's motherboard. Hilodog is a sparc IIi. I forgot to remove hammer's P2 IO board from behind the VME bus. Power the system, immediately smelled smoke, and powered down the system. Power was on for about 5-7 seconds. Before doing any testing on the motherboards, I did unseated the DSPs/Fiber Interface board from the VME backplane. I was trying to test hilodog (our sparc IIi, similar to bigdog's motherboard). This was damaged during the a Nov2011 Hilo electrical storm. It was booting and freezeing in Nov, but last week I unable to get it to boot up (screen always black). I think these are true, but I am not that familiar with VME hardware: P2 connect on the VME blackplane is not part of the VME bus (the bus interconnecting the all the boards in the chassis). P2 pin-outs are specified to each VME board, usually used for additional connector (daughter board are very common). I looked at the manual for the SPARC 5/65 & USPIIi The 5/64 defines J2 C1 as GND. The USPIIi defines it as VCC. I likely VCC & GND on the USPIIi Motherboard via the P2,5/64's bread board. The USPIIi is likely dead. 1218-pics ToDO 0. We need make a boot a backup boot disk. Have problem with using miranda's old sun disk. Don't have a working VME develop system backup to make or test the backup. 20 dec - purchased harddisk. 1. Check vendors for Themis CPU. Themis USPIIi-1v (UltraSPARC-IIi 270MHz, 128MB RAM), Solaris 2.6 www.artisan-scientific.com Still looking. 2. network booting. Look into network booting of the sparc stations.