The tag has no wiki summary.

learn more… | top users | synonyms

-1
votes
3answers
3k views

iostat shows 100% utilization - what can be done to improve performance?

I have dedicated web server with CentOS. Recently, I get very high server load, probably caused due to IO bottleneck. iostat -mx 3 shows: avg-cpu: %user %nice %system %iowait %steal %idle ...
2
votes
2answers
317 views

How do I know if my disks are being hit with too many I/O reads or writes or both?

I know a bit about disk I/O and bottlenecks relating to this especially when relating to databases. How do I really know what the max I/O numbers will be for my disks? What metric might be available ...
0
votes
1answer
137 views

Get I/O TPS for NFS on Solaris machines

I'm attempting to get IOPS for my Solaris servers. Typically I use tps for this calculation, however: iostat gives TPS but only for local disks. iostat -x gives everything but TPS for ALL disks. sar ...
0
votes
1answer
310 views

High cancelled_write_bytes, high IO load, concurrent memory-mapping

I have a software for persisting/retrieving data from files (a bit like database, but it's not exactly a database engine). It's not open source and the documentation is sparse, therefore I don't have ...
0
votes
1answer
82 views

how can I tell if there is a cpu trashing on my linux machine

I want to write a script that will kill my processes when my machine is doing "too much" swapping of process memory from disk to memory. What is the best way to determine that? I saw top has %wa and ...
0
votes
1answer
679 views

SNMP monitoring of iostat on a Solaris 10 server?

I'm looking at the System Management Agent resources which gives guidance to the Solaris SNMP tools however after extensive googling I can't find any OID reference list or details on what can be ...
1
vote
0answers
350 views

MySql/RedHat, High I/O Wait … for a few hours… then disappears

I've got a RH server as a VMWare instance logging GPS fixes 24x7. The transaction volume hasn't changed over the months (in fact I am pruning the transaction table more aggressively now) with only ...
0
votes
0answers
46 views

Strange load average behaviour

We have a daemon which collects system data and writes it into database. Sometimes the load average goes very high without apparent reason. Here is sample output. | id | server_id | datetime ...
0
votes
0answers
101 views

confused by iostat reporting striped mdadm more tps than aggregate volumes

I've got an mdadm striped device (/dev/md0) from /dev/sdm{1..4}. Here is an output of iostat 30 and then mdadm --detail on the mdadm device # iostat 30 Linux 2.6.32-309-ec2 (ip-xx-xx-xx-xx) ...