PLEASE DO NOT REMOVE ANY ITEM FROM THIS LIST. IF IT IS COMPLETE, THEN PLEASE MOVE IT TO THE END OF THE LIST UNDER "DONE".
TO DO
Order fan trays and install:
Rittal solution does not work, TRX temp is down 1K (from 44C to 43C)
Need second prototype, might have long lead time, waiting for reply
Order inrush current limiters and install:
2 ordered to see if they work and which one is sufficient, delivery: week after Easter
Get proper mesh installed on top of power distribution units
GTE claims that work has been done which isn't, CA took pictures and sent them to Stegmann
Interlock for
UPS - right now there is only a lock, no interlock:
?
Absorbing foam on ceiling:
?
metal panel covering the power cables between
UPS racks:
?
Split the load evenly between 3 power phases (we are back where L1 > L2 ~ L3):
?
Metering for in-room AC units to show 100% instead of 66%:
Company said that's not going to happen, 100% is gauged to design air pressure, locally values might deviate
Ventilated floor tiles need to be moved, room corners should get these to improve air flow :
Filter in primary cooling circuit present?:
?
- rd level temp protection:
Electronic locks for cluster room:
Klaus Haupt is working on it
Student's room:
Streich is notified that room is empty, nothing has happened yet.
pump room door does not close properly: :
humidity control does not work in cluster room:
Hardware
Get CX-4 cards for Suns and test:
12 cards should arrive 2008-03-13 at Pyramid, should arrive @AEI 14th or 17th.
avnet just contacted me, the cards will leave the US this week
Get CX-4 cables for Suns :
Ordered, ETA week after Easter
Order new 4000 series card for Cisco chassis:
Ordered, first ETA BEGINNING OF JUNE
computacenter.com is trying to get one sooner
Get Sun boxes installed:
Moved to March 14th (hopefully 18th), no strong students around
Get Linux server boxes installed:
Most are installed, need to get new kernel stress tested
There seems to be an off by one error in MAC addresses
New ETA: March 19th
Get NUT properly configured and working:
Ran into problems with more than 900 nodes
Will resolve by using 2 upsd servers
SNMP monitoring of RAID cards for failed disks or problems:
Should be available March 19th or 20th
Decisions/Testing/External help
SNMP monitoring of network switch for failed power supplies:
Need to ask Max
SNMP monitoring of network switch for problems (failed line or fabric cards):
Need to ask Max
Get extra managment module and power supplies installed in EFX-1000 switch:
Need to ask Max (at least for PSU)
Find price (and perhaps order) two additional power supplies for EFX-1000 switch:
Asked Rischau, waiting for reply
Get USB memory card (if needed) for second EFX-1000 management card:
Need to ask Max
Neterion Card for Linux file servers:
Arrived 2008-03-07, needs to be tested
Get SMS warning messages for overtemperature and failure conditions working:
Sending SMS works, only little scripting needed
DONE
Get 6ft networking cables:
Ordered, arrived 2008-03-10
Get GPS antenna and NTP clock installed and working:
Up and running, IP: 10.20.40.10
Exchange parts for cluster (from Pyramid):
soon, should all have been in Freiburg already, Rischau is checking