This page is intended to supplement material in the data cube documentation where we have had fixable issues 🙂

These fixes are implemented using our own telemetry box and scripts.

Changing the parameters on the DataCube remotely

We can change the parameters on the DataCube by logging in remotely via a telemetry box.

To access the filesystem, ssh into ssh.geos -> stream -> telemetryBox then run the mount-cube.sh script.

The parameters are reset when the cube is unmounted.

A pdf of all configuration options can be found here. The basic options include:

E_NAME= Experiment Name
S_RATE= Samplerate
F_TIME= Flush SD time und GPS cycletime [min]
GPS_TI= GPS on-time [min]
GPS_OF= GPS off after / 0=GPS_TI 1=60 valid GPS strings
A_FILT= Digital Filter 0=HPFoff 1=HPFon
A_PHAS= Filter characteristics 0=LinPhase 1=MinPhase
GPS_ON= GPS Option 0=cycled 1=continuously
PLL_XO= OszillatorKorrektur / 1=PLL 2=DIFF
GPS_PO= if set, then initial position
P_AMPL= factor of amplification (P_AMPL * RefVolt/24bit) (from Config File) 
C_MODE= ADC Mode / 0=LowPwr 1=High Res(from Config File)
A_CHOP= Amplifier Chopping / 0=off 1=on (from Config File)
CH_NUM=number of active channels(from Config File)

Fixing Corrupted Storage on Datacube

Sometimes the storage becomes corrupted – we have not diagnosed why…

The corrupt storage prevents old files being deleted which causes the cube to fill up.

This can be fixed remotely using the telemetry box.

Symptoms:

Firstly, we need to mount the data cube so that we can use the tools below on linux or mac to analyse it.

and

pi@raspberrypi:/media/datacube-memory $ dmesg
[  323.463797] usb 1-1.3: new high-speed USB device number 7 using dwc_otg
[  323.600488] usb 1-1.3: New USB device found, idVendor=0424, idProduct=2240, bcdDevice= 1.98
[  323.600534] usb 1-1.3: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[  323.600558] usb 1-1.3: Product: Ultra Fast Media
[  323.600578] usb 1-1.3: Manufacturer: Generic
[  323.600599] usb 1-1.3: SerialNumber: 000000225001
[  323.617176] usb-storage 1-1.3:1.0: USB Mass Storage device detected
[  323.624830] scsi host1: usb-storage 1-1.3:1.0
[  324.645307] scsi 1:0:0:0: Direct-Access     Generic  Ultra HS-COMBO   1.98 PQ: 0 ANSI: 0
[  324.648533] sd 1:0:0:0: [sdb] 62333952 512-byte logical blocks: (31.9 GB/29.7 GiB)
[  324.649735] sd 1:0:0:0: [sdb] Write Protect is off
[  324.649776] sd 1:0:0:0: [sdb] Mode Sense: 23 00 00 00
[  324.650923] sd 1:0:0:0: [sdb] No Caching mode page found
[  324.650958] sd 1:0:0:0: [sdb] Assuming drive cache: write through
[  324.667269]  sdb: sdb1
[  324.672207] sd 1:0:0:0: Attached scsi generic sg1 type 0
[  324.672969] sd 1:0:0:0: [sdb] Attached SCSI removable disk
[  334.157443] FAT-fs (sdb1): Volume was not properly unmounted. Some data may be corrupt. Please run fsck.
[  766.938540] FAT-fs (sdb1): error, fat_get_cluster: invalid start cluster (i_pos 0, start 07780080)
[  766.938577] FAT-fs (sdb1): Filesystem has been set read-only
[  766.938781] FAT-fs (sdb1): error, fat_get_cluster: invalid start cluster (i_pos 0, start 07780080)
[  766.952773] FAT-fs (sdb1): error, fat_get_cluster: invalid start cluster (i_pos 0, start 045f0080)
[  766.953041] FAT-fs (sdb1): error, fat_get_cluster: invalid start cluster (i_pos 0, start 045f0080)
[  766.965087] FAT-fs (sdb1): error, fat_get_cluster: invalid start cluster (i_pos 0, start 01570080)
[  766.965406] FAT-fs (sdb1): error, fat_get_cluster: invalid start cluster (i_pos 0, start 01570080)
[  768.831262] FAT-fs (sdb1): error, fat_get_cluster: invalid start cluster (i_pos 0, start 32303030)
[  768.831526] FAT-fs (sdb1): error, fat_get_cluster: invalid start cluster (i_pos 0, start 32303030)
[  768.841817] FAT-fs (sdb1): error, fat_get_cluster: invalid start cluster (i_pos 0, start 1bb00600)

Solution

First the cube was mounted by ssh into telemetry box and then running mount-cube.sh. Followed by sudo umount /media/datacube-memorybecause fsck needs the datacube powered but not fully mounted.

This can also be achieved using the grey cable from a mac or linux machine because they have fsck.

Auto repair with fsck did not to it. Calum used:sudo fsck -vck /dev/sdb1

This produced:

/210812
Directory has non-zero size. Fixing it.
/210812
Has a large number of bad entries. (764/779)
Drop directory ? (y/n) y
Checking for unused clusters.
Reclaimed 862993 unused clusters (28278554624 bytes).
Checking free cluster summary.
Free cluster summary wrong (54729 vs. really 917811)
1) Correct
2) Don't correct
? 1
Perform changes ? (y/n) y
/dev/sdb1: 2428 files, 55773/973584 clusters

It then reported 6% full rather than 96%.

Share