Skip to Content.
Sympa Menu

perfsonar-user - Re: [perfsonar-user] About perfSONAR toolkits 3.4.2 Throughput / Latency Graphs problem

Subject: perfSONAR User Q&A and Other Discussion

List archive

Re: [perfsonar-user] About perfSONAR toolkits 3.4.2 Throughput / Latency Graphs problem


Chronological Thread 
  • From: "Andrew Lake" <>
  • To: "Zhiyan Zheng" <>
  • Cc: "GarnizovIvan (RRZE)" <>,
  • Subject: Re: [perfsonar-user] About perfSONAR toolkits 3.4.2 Throughput / Latency Graphs problem
  • Date: Fri, 07 Aug 2015 05:21:21 -0700 (PDT)

Hi,

The heap space is the problem. How much memory does your host have? Also a common mistake we’ve seen people make is they install the 32 bit ISO (i386/i686) on a 64-bit machine (x86_64). When that happens, your system can only use 4GB of memory regardless if more than that is actually installed, which can confuse software like cassandra which sets its heap size based on a percentage of the memory installed. The command ‘uname -a’ should tell you the version you have running. 

Thanks,
Andy




On Thu, Aug 6, 2015 at 10:10 PM, Zhiyan Zheng <> wrote:

Hi Andy and Ivan,

Thanks for your kind replies,

# ls -alt /var/run/cassandra/cassandra.pid 

-rw-r--r-- 1 cassandra cassandra 5 Aug  7 10:00 /var/run/cassandra/cassandra.pid


# more /var/log/cassandra/cassandra.log 

Error occurred during initialization of VM

Could not reserve enough space for object heap

Error: Could not create the Java Virtual Machine.

Error: A fatal exception has occurred. Program will exit.


# more /var/log/cassandra/system.log

/var/log/cassandra/system.log: No such file or directory


Thanks,
Zhiyan Zheng
From: Andrew Lake <>
Date: Thursday, August 6, 2015 at 6:12 PM
To: "GarnizovIvan (RRZE)" <>
Cc: <>, Zhiyan Zheng <>
Subject: RE: [perfsonar-user] About perfSONAR toolkits 3.4.2 Throughput / Latency Graphs problem

Hi,

The problem actually appears to be that cassandra is dying shortly after restart. Touching the pid is likely not going to help. A restart is just a “stop” then a “start”. The stop is failing because cassandra is already down so it the proceeds to try and start it. Could you send /var/log/cassandra/cassandra.log and /var/log/cassandra/system.log?

Thanks,
Andy




On Thu, Aug 6, 2015 at 1:20 AM, Garnizov, Ivan (RRZE) <> wrote:

Hi Zhiyan,

 

Nice to have you back and hope you are doing well.

For this issue, please touch the folder and the file /var/run/cassandra/cassandra.pid  with for owner cassandra:cassandra and make it writable only by the owner.

 

Please also point us to the netinstall image you used!

 

Best regards,

Ivan

 

From: [] On Behalf Of zhzhy
Sent: Donnerstag, 6. August 2015 04:55
To:
Subject: [perfsonar-user] About perfSONAR toolkits 3.4.2 Throughput / Latency Graphs problem

 

Hi all

I installed PS toolkits 3.4.2 by netinstall ISO image, when I add a new schedule test, there is nothing for the result, I got Error retrieving test data: undefined.

And when I try to restart the services of cassandra, I got the messages as:

# service cassandra restart

Shutdown Cassandra: bash: line 0: kill: (12239) - No such process

ERROR: could not stop cassandra: cassandra dead but pid file exists

Starting Cassandra: OK

# service cassandra status

cassandra dead but pid file exists



# service cassandra restart

Shutdown Cassandra: cat: /var/run/cassandra/cassandra.pid: No such file or directory

kill: usage: kill [-s sigspec | -n signum | -sigspec] pid | jobspec ... or kill -l [sigspec]

ERROR: could not stop cassandra: cassandra dead but subsys locked

Do anybody have idea about this problem?

Best Regards,

Zhiyan Zheng






Archive powered by MHonArc 2.6.16.

Top of Page