pt-stalk usage questions

[krishna]

Hi Team,

Could anyone give me more examples wrt pt-stalk with examples and options like --cycles --deamonize --function --match --threshold --disc-pct-free —retention-time including the output would be more clear. I have been through the link http://www.percona.com/doc/percona-toolkit/2.2/pt-stalk.html but require more clarification with examples. Please Help!!

Best Regards,
Krishna

[krishna]

Hi Tom,

Thanks for the detailed explanation that is being shared, the recorded webinar was wonderful. However, would like to know more about pt-stalk --threshold information. The information available in the blog is not enough for me. I would need more examples such that I can check as below using pt-stalk:

1. What is the importance of threshold.
2. When threshold is given as 10, does that mean 10 connections are opened or collect the information keeping in view when 10 threads are running.
3. In general the default value of threshold is 25 or how much.
4. What does connections imply in this case here. Is it connections to DB or what?
5. How can we best use the option threshold to capture the average throughput, let me know if there is any other option through which we can get them.
6. Can we get the details related to dashboard, ganglia and mysql services running using any of the options in pt-stalk please specify.

Please suggest!

Best Regards,
Krishna

[krishna]

Thanks Tom,

Been through the documentation as I mentioned in the first email, but need more clarity on that. Hence, look forward to hear from PT Toolkit expert to help me find answer to my queries.

Best Regards,
KP

[mirfan]

Let me show you couple of examples of pt-stalk
sudo ./pt-stalk --no-stalk --iterations=2 --sleep=30 --dest=$PTDEST – --user=root --pass=;

Here --no-stalk will start collecting data immediately and --iterations tells how many times to collect diagnostic data while --sleep tells how long to sleep (in seconds) between each data collection cycle and --daemonize will run the job in background continuously and start collecting data when pt-stalk triggers on specified condition. You will find full documentation in pt-stalk doc here http://www.percona.com/doc/percona-toolkit/2.2/pt-stalk.html

To get an example of --function along pt-stalk you will find good example of it in my post here http://www.percona.com/blog/2014/05/02/how-to-identify-and-cure-mysql-replication-slave-lag/

Hope that helps.

[krishna]

Thanks Irfan, It did help.