how to check for "No space left on device"

[Pila]

My Asus RT-AC68U regularily develops a lack of free memory, often 80+ days. Now it did so in about 40 days likely because I was trying to do something and was experimetning with some settings changes. It stopped accepting my changes (Port Forward). My script (DDNS) would not complete its task as normal.

Log shows:

Jul 21 23:58:30 dnsmasq-dhcp[432]: failed to write /var/lib/misc/dnsmasq.leases: No space left on device (retry in 60s)

Tools tab does not show any memory filled up! Never! I do not care about putting a newer fw version in a hope of a fix. I have seen this topic numerous times, but nothing good.

I would just like to be able to notice this problem from a script so it can reboot a router automatically. Anybody knows how could I diagnose such a problem from a script, aside from parsing of the log? Anybody knows jow to stop this problem from occuring?

 

[thelonelycoder]

My Asus RT-AC68U regularily develops a lack of free memory, often 80+ days. Now it did so in about 40 days likely because I was trying to do something and was experimetning with some settings changes. It stopped accepting my changes (Port Forward). My script (DDNS) would not complete its task as normal.

Log shows:

Jul 21 23:58:30 dnsmasq-dhcp[432]: failed to write /var/lib/misc/dnsmasq.leases: No space left on device (retry in 60s)

Tools tab does not show any memory filled up! Never! I do not care about putting a newer fw version in a hope of a fix. I have seen this topic numerous times, but nothing good.

I would just like to be able to notice this problem from a script so it can reboot a router automatically. Anybody knows how could I diagnose such a problem from a script, aside from parsing of the log? Anybody knows jow to stop this problem from occuring?

How big is that file?
There ought to be plenty of free space in the /tmp/ directory. I believe this file uses the same free space as /tmp/ uses.

 

[john9527]

I would just like to be able to notice this problem from a script so it can reboot a router automatically. Anybody knows how could I diagnose such a problem from a script, aside from parsing of the log? Anybody knows jow to stop this problem from occuring?

run

df /tmp

When it fills up, causing your problem, your need to hunt down the file that's growing unbounded (may be a runaway log). Once you identify what file or files are causing the problem, we may be able to offer some advice.

 

[Pila]

run

df /tmp

When it fills up, causing your problem, your need to hunt down the file that's growing unbounded (may be

OK, that sounds like a plan I think the problem may be connected with me experimenting something and doing a lot (20+) of GUI changes (e.g. port forward off / on in the last case). Zhat may also explain irregularity of the problem as it pops now and there so I keep forgeting about it and have never hunted it down.

 

[Pila]

Another thing which may be relevant: sometimes it takes a long time for putty to connect to the router. It may take it 30 seconds ormore, maybe even more to connect to the router. Screen is just blank.

In the meantime, I can use the same putty and the same computer and network to connect instantly to 5 other Linux devices, so the problem exists only with connection to my asus router.

It just refused me, once I got netowrk timeout from my local PC, no log entry at all within my reouter. Second time I was unable to puty in manually entering ssh pwd, log said:

Jul 24 11:37:00 dropbear[29570]: Child connection from 192.168.1.90:54142
Jul 24 11:37:21 dropbear[29570]: Exit before auth: Error reading: Connection timed out

Then my router reverts back to normal instant logins for no apparent reasons. There are no large loads killing router, nothing changes much there.

 

[Pila]

I think I may be nar the same memory problem as before. But, I can find no way to get the number of free memory that the router is displaying.

After a reboot, my router says (Tools menu):
Total 249.71 MB
Free 194.28 MB

Currently it says:
Total 249.71 MB
Free 91.29 MB

This free amount seems to be a problem treshold. But:

# df -m /tmp
Filesystem           1M-blocks      Used Available Use% Mounted on
tmpfs                      125         1       124   1% /tmp

So obviously that 91 MB is not related to the /tmp.

So: where is this 91 mb free (down from 194 MB just after a reboot) and how do I check it from the command line?

Seems like top may be of use:

Mem: 163904K used, 91796K free, 0K shrd, 1968K buff, 9688K cached
CPU:  0.0% usr  0.0% sys  0.0% nic 92.8% idle  0.0% io  0.0% irq  7.1% sirq
Load average: 0.02 0.08 0.06 1/85 14024
  PID  PPID USER     STAT   VSZ %VSZ CPU %CPU COMMAND
 1140   445 pila     S    92576 36.1   0  0.0 networkmap

How can I reboot the networkmap as it seems to use 36% of memory (it seems like so to me). Can i just kill this process and restart it?

# ps | grep ma[p]
 1140 pila     92576 S    networkmap

Seems like killing it did not do any harm. For now. Must I manually rerun it after killing it or can I just let it alone? Seems like it does not finish when I run it manually (it was stil left open after 30 minutes)

Is there any regular way to monitor and reset/reboot this process? Or must I cru regular check of this process, and when it is over desired limit, kill it and rerun it?