RT-86U - USB 2.0 port not mounting at boot.

[ColinTaylor]

So it should now be recording the information every 10 minutes. To see today's disk information use "sar -dh".

# sar -dh
Linux 2.6.36.4brcmarm (RT-AC68U)        06/22/20        _armv7l_        (2 CPU)

00:00:00          tps     rkB/s     wkB/s     dkB/s   areq-sz    aqu-sz     await     %util DEV
00:10:00         0.00      0.0k      0.0k      0.0k      0.0k      0.00      0.00      0.0% mtdblock3
00:10:00         0.02      0.0k      0.2k      0.0k      7.4k      0.00      3.08      0.0% sda
00:20:00         0.00      0.0k      0.0k      0.0k      0.0k      0.00      0.00      0.0% mtdblock3
00:20:00         0.02      0.0k      0.1k      0.0k      5.2k      0.00      5.00      0.0% sda
00:30:00         0.00      0.0k      0.0k      0.0k      0.0k      0.00      0.00      0.0% mtdblock3
00:30:00         0.03      0.0k      0.1k      0.0k      5.6k      0.00      2.00      0.0% sda
00:40:00         0.00      0.0k      0.0k      0.0k      0.0k      0.00      0.00      0.0% mtdblock3
00:40:00         0.03      0.0k      0.1k      0.0k      5.6k      0.00      2.67      0.0% sda
:
:
13:10:00         0.00      0.0k      0.0k      0.0k      0.0k      0.00      0.00      0.0% mtdblock3
13:10:00         0.06      0.8k      0.2k      0.0k     18.0k      0.00      4.71      0.0% sda
Average:         0.00      0.0k      0.0k      0.0k      0.0k      0.00      0.00      0.0% mtdblock3
Average:         0.03      0.0k      0.1k      0.0k      6.0k      0.00     11.12      0.0% sda

sar saves information for each day of the last 31 days. So for example if I wanted to look at data for the last "day 19" (i.e. 19th June) I would use:

sar -dh -f /opt/var/log/sysstat/sa19

 

[miazza]

Nice

miazza@RT-AC86U-3150:/tmp/home/root# sar -dh
Linux 4.1.27 (RT-AC86U-3150)    06/22/20        _aarch64_       (2 CPU)

13:50:47     LINUX RESTART      (2 CPU)

14:00:00          tps     rkB/s     wkB/s     dkB/s   areq-sz    aqu-sz     await     %util DEV
14:10:00         0.00      0.0k      0.0k      0.0k      0.0k      0.00      0.00      0.0% sda
14:10:00         0.60      4.1k      4.7k      0.0k     14.6k      0.14    225.65      4.0% sdb
14:20:00         0.00      0.0k      0.0k      0.0k      0.0k      0.00      0.00      0.0% sda
14:20:00         0.47      0.0k      3.9k      0.0k      8.5k      0.11    230.11      3.6% sdb
Average:         0.00      0.0k      0.0k      0.0k      0.0k      0.00      0.00      0.0% sda
Average:         0.53      2.0k      4.3k      0.0k     11.9k      0.12    227.59      3.8% sdb

Next step is to uderstand why there is an average 3,8% on sdb ...

 

[ColinTaylor]

Without any parameters sar will report CPU usage. Another more useful option is to report swapping activity. In normal usage you want no swapping occurring, with only some occasional activity overnight when housekeeping tasks are running.

sar -W

EDIT: I have discovered that the kernel VM event counters (CONFIG_VM_EVENT_COUNTERS) are not enabled on the router . So sar options -W (swapping) and -B (paging) will always report 0 (see the post below).

 

[miazza]

Paging and CPU looks normal:

sar -B
Linux 4.1.27 (RT-AC86U-3150)    06/22/20        _aarch64_       (2 CPU)

13:50:47     LINUX RESTART      (2 CPU)

14:00:00     pgpgin/s pgpgout/s   fault/s  majflt/s  pgfree/s pgscank/s pgscand/s pgsteal/s    %vmeff
14:10:00         0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00
14:20:00         0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00
14:30:00         0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00
14:40:00         0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00
14:50:00         0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00
Average:         0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00      0.00
miazza@RT-AC86U-3150:/tmp/home/root# sar
Linux 4.1.27 (RT-AC86U-3150)    06/22/20        _aarch64_       (2 CPU)

13:50:47     LINUX RESTART      (2 CPU)

14:00:00        CPU     %user     %nice   %system   %iowait    %steal     %idle
14:10:00        all      1.70      0.00      1.65      0.60      0.00     96.05
14:20:00        all      1.10      0.00      1.09      0.35      0.00     97.46
14:30:00        all      1.17      0.00      1.35      0.14      0.00     97.34
14:40:00        all      0.97      0.00      1.39      0.31      0.00     97.33
14:50:00        all      0.86      0.00      1.19      0.29      0.00     97.66
Average:        all      1.16      0.00      1.33      0.34      0.00     97.17

 

[miazza]

Here few pastebin with data collection.
https://pastebin.com/GUKiszKz
https://pastebin.com/qqYs9sBD
https://pastebin.com/NpevwXbA

As far as I can see there is something writing constantly on sdb ... not big size but constant.

Now I have to understand what is the process that is doing that; my be some script log ...

 

[ColinTaylor]

@miazza See my edit in post #51 regarding sa2.

As far as I can see there is something writing constantly on sdb ... not big size but constant.

Now I have to understand what is the process that is doing that; my be some script log ...

How many partitions does sdb have?

Try using this as a starting point:

fuser -mv /dev/sdb1

If you don't have the fuser command you'll need to install it.

opkg install psmisc

 

[miazza]

Thanks for the advise at post #51. I will delete when back home.
sdb has one single ext4 partition.

I have installed psmisc (acting from mobile via vpn).

I cannot cut and paste from mobile but fuser command does not show anything strange:

- myswap.swp
- Entware mount
- pixelserv-tls
- dnsmasq

Just by comparison, it would be nice to compare this output with another user with a standar install like mine.
So far I don't see anything that could justify the killing of my drives... may be I have been extremely unlucky or all my "gift drives" were really of poor quality.