Nlbwmon hangs after few days of use

Oh, I forgot to mention that I survived 40 days of uptime and nlbwmon was still counting just before reboot due to updates.

I don't understand how your config list local_network 'lan' follows this

Hi All,

OK, few things. If your nlbwmon is crashing, first thing to do is see what parameters were passed to start it.

ps | grep nlbwmon

Look at the -s parameter, if you have multiple monitored networks with overlapping address spaces you should remove list local_network 'lan' from /etc/config/nlbwmon

As of today I'm at 129 days uptime, and still functional nlbwmon. My router is on a UPS, and yeah, I need to patch. This is my config:

cat /etc/config/nlbwmon

config nlbwmon
        option netlink_buffer_size '1048576'
        option database_interval '1'
        option database_limit '10000'
        option protocol_database '/usr/share/nlbwmon/protocols'
        option database_generations '24'
        option database_directory '/mnt/sda1/nlbwmon'
        option commit_interval '4h'
        option refresh_interval '1m'
        list local_network '192.168.1.0/24'

cat /etc/sysctl.d/12-nlbwmon.conf

net.core.rmem_default=1048576
net.core.wmem_default=1048576
net.core.rmem_max=1048576
net.core.wmem_max=1048576

ps | grep nlb

12378 root      1456 SN   /usr/sbin/nlbwmon -o /mnt/sda1/nlbwmon -b 1048576 -i 4h -r 1m -p /usr/share/nlbwmon/protocols -G 24 -I 1 -L 10000 -Z -s 192.168.1.0/24
1 Like

After 157 days of uptime, my nlbwmon has crashed :frowning:

Fri May  5 08:11:11 2023 daemon.err nlbwmon[12378]: Netlink receive failure: Out of memory
Fri May  5 08:11:11 2023 daemon.err nlbwmon[12378]: Unable to dump conntrack: No buffer space available
Fri May  5 17:13:54 2023 daemon.err nlbwmon[12378]: Netlink receive failure: Object busy
Fri May  5 17:13:54 2023 daemon.err nlbwmon[12378]: Unable to dump conntrack: I/O error

I suppose I should patch, then reboot at least once a month. Ug

1 Like

Patched to 22.03.5 on July 17.
Tried to access stats on August 22, found nlbwmon is once again broke. Ugh.

Throwing in the towel and scheduling a reboot on day 1 of each month :frowning:

1 Like

Thanks for posting all your findings so far. I increased the memory limit/buffer size to 5MB - will report back if this changes anything.

My logs worked fine yesterday as they have for several months now. Writing to ext4 on external USB. Today the log for the month is 0 bytes. I just looked at it a few days ago and it was full of data. There's still something amiss.