I am having a problem with for some reason simple queues not working correctly.

I have approx 320 simple queues, all with varying attributes. I have chosen this method above all others for numerous reason so please don't just tell me to use PCQ, PPPoE, PPTP, or anything else.

Basicly what happens is (and this is all the sudden) in the properties for the simple queue, the 'total statistics' tab will record the actually data rate (closely matches a torch run), but the 'statistics' has very low data rate(s).

Ex.
'Total Statistics' Total Avg Rate: 3.8 Mbps
'Statistics' Target Upload: 51.5 kbps
'Statistics' Target Download: 1984 bytes

As a result of the simple queue 'not seeing' the data properly it doesn't limit it to the appropiate limits. Recreating the queue has the same effect, and other simple queues are operating fine.

This is a very annoying problem as I typically trust that RouterOS is doing its job. I caught this by mistake, its not something that is easily noticed.

Any idea what is causing this?

MyThoughts -
Boy that is a lot of queues..... :)

I would bet money you are using a PC based ROS distro, and you're using ROS 3.12 or ROS 3.13. From what I have seen on my ROS 3.13 router it acts similiar to yours. Look at your mangle rules and you'll notice that they are probably not counting corectly either, and so the 'simple' queues do not work correctly either (and neither do PCQ, SFQ or any other queing scheme)...especially if you have 'default-small' as the queue type.... I have ONLY seen this on the PC based version - looks like the RB based version of ROS is unaffected in this regard.

I have sent a supout.rif file to MT, got a few comments back but no real answers. I am waiting for A) a real answer that includes a fix or viable workaround OR B) ROS 3.14 which MT alluded would indeed 'fix' the issue....

My advice, send them a supout.rif and complain like I did - that way there will be one more voice out there telling MT that there is a big issue that needs to get fixed....

R/

I am too having issues with simple queues after upgrade from 2.9.xx to 3.13. (Actually went to 3.14 but AMD dual core issue with clock, and downgraded back to 3.13)

We use data from simple queues for each customer IP to bandwidth stats.

I've got at least one queue for a customer's IP that will seem to stick at a RX / TX Avg, and the totals don't change. Sometimes a reset gets it going.

This morning I noticed the queue had blank for total TX bytes.

I've just deleted the problem one and then recreated and it still won't work right. Now, it's showing Rx avg, Total Rx, Total Tx, but blank for tx avg.

The total bytes are all blank after upgrade

It may have started when we took out a customers router and put in an RB450 for them.

Here's a current print stats for the queue in question.
211 target-addresses=xx.xx.xx.xx/32 dst-address=0.0.0.0/0 rate=4776/0
total-rate=0 packet-rate=4/0 total-packet-rate=0 queued-bytes=0/0
total-queued-bytes=0 queued-packets=0/0 total-queued-packets=0
bytes=3437635/1389884 total-bytes=0 packets=12495/3287 total-packets=0
dropped=0/0 total-dropped=0 overlimits=0/0 total-overlimits=0
lends=411/111 total-lends=0 borrows=12084/3176 total-borrows=0
pcq-queues=0/0 total-pcq-queues=0

Thanks,
Randall

ver 3.14

Simple queue 256Kbps/1Mbps on pptp connection

Statistics:
Target upload 234.4 kbps, Target download 0kbps
Torch shows upload 234 Kbps, download 12Mbps!

Some other simple queues on pptp work fine.
Thinking downgrade ver 2