Search/Navigation:
Related:
SLAC/EPP
/HPS Public
Jefferson Lab/Hall B
/HPS Run Wiki
S30XL-LESA/LDMX
...
There should be multiple partitions available for mounting.
Code Block |
---|
RTEMS RTEMSAPP LINUX SCRATCH BOOT |
Mount the first "boot" partition. E.g.:
...
Code Block |
---|
cp dpm11.env /mnt/sdcard/uboot.env cp dpm11.env /mnt/sdcard/uboot_linux.env |
Figure out
Output from crate front panel screen:
“not booted” is fine, we are looking for “not ready”
dpmp1-rce2 - not ready => this is dpm11.env ?
We are replacing sd card 11 (rce 2 on card dpm card 1)
After the replacement. Same problem.
Pull card out, make sure it’s fully pushed in.
Same problem.
Copied an existing dpm fpga.bit file (from dpm0) onto the sd card. Noted that the existing one on the SD card was 4 bytes larger than the one running on dpm1. Will save the existing fpga.bit somewhere on Rafo’s laptop.
Plugged SD card into a good now dpm1/rce0 and dpm1/rce2 are "not_ready”. The problem follows the SD card.
Going back up to put it into a computer we can access remotely.
Problem was
$ dd if=/home.orig/hpsrun/hps_dpm.bin of=/dev/sdb1 bs=1M
vs (correct)
$ dd if=/home.orig/hpsrun/hps_dpm.bin of=/dev/sdb bs=1M
Then it worked.
Actions:
Prepare full range of SD cards to be swapped into the cobs that had two failures (top one)
Turn off logging to file system, ryan will help to turn off journaling(?) and I need to turn off logging redirect of the servers. Perhaps log to NFS?
COB hot plugging:
...
out which RCE the newly created SD card correspond to: see page.
COB Startup Checks
COB status:
Identify RCEs from the front panel by noting the cycling patterns that are printed.
We are looking for “not ready”, e.g. "dpmp1-rce2 - not ready".
...
RTM status: