10-11pm RRD gap in graphs (4.1.1)
list Craig Cook
I have an odd issue... From around 10pm at night to around 11pm (EST) I have a gap in my CPU and Processes/Users graphs. I haven't seen any reasons in the hobbit logs. Other RRD graphs are fine (disk, vmstat, etc) This is not just one night, but every night. I am using the hobbit client on the hobbit server, and hobbit client on another linux box (both are CentOS 4). Has anyone else seen this? What is significant about CPU, Proc/Users graphs? Thanks Craig Cook -- Systems Monitoring Consulting and Support Services http://www.cookitservices.com
list Robert Taylor
I actually get the same thing, but it happens between 5:30 and 6:30 every day. Robert
▸
-----Original Message----- From: Craig Cook [mailto:user-618593604956@xymon.invalid] Sent: Monday, September 19, 2005 8:53 PM To: user-ae9b8668bcde@xymon.invalid Subject: [hobbit] 10-11pm RRD gap in graphs (4.1.1) I have an odd issue... From around 10pm at night to around 11pm (EST) I have a gap in my CPU and Processes/Users graphs. I haven't seen any reasons in the hobbit logs. Other RRD graphs are fine (disk, vmstat, etc) This is not just one night, but every night. I am using the hobbit client on the hobbit server, and hobbit client on another linux box (both are CentOS 4). Has anyone else seen this? What is significant about CPU, Proc/Users graphs? Thanks Craig Cook -- Systems Monitoring Consulting and Support Services http://www.cookitservices.com
list Henrik Størner
▸
On Mon, Sep 19, 2005 at 07:52:43PM -0500, Craig Cook wrote:
I have an odd issue...From around 10pm at night to around 11pm (EST) I have a gap in my CPU and Processes/Users graphs.I haven't seen any reasons in the hobbit logs. Other RRD graphs are fine (disk, vmstat, etc) This is not just one night, but every night. I am using the hobbit client on the hobbit server, and hobbit client on another linux box (both are CentOS 4). Has anyone else seen this? What is significant about CPU, Proc/Users graphs?
It's a bug in the client handler. The reason why it happens for one hour every day is that the "uptime" output is different for that hour. Most of the time, it looks like 08:26:57 up 4 days, 10:22, 3 users, load average: 0.01, 0.10, 0.06 but for one hour each day - starting N*24 hours after it was booted - it will be 08:26:57 up 7 days, 13 mins, 3 users, load average: 0.01, 0.10, 0.06 which the 4.1.1 version of hobbitd_client fails to recognize. I really need to get 4.1.2 finished ... Henrik
list Etienne Roulland
Hello, same thing from some clients from 6 to 7pm... Etienne
▸
I actually get the same thing, but it happens between 5:30 and 6:30 every day. Robert -----Original Message----- From: Craig Cook [mailto:user-618593604956@xymon.invalid] Sent: Monday, September 19, 2005 8:53 PM To: user-ae9b8668bcde@xymon.invalid Subject: [hobbit] 10-11pm RRD gap in graphs (4.1.1) I have an odd issue...From around 10pm at night to around 11pm (EST) I have a gap in my CPU andProcesses/Users graphs. I haven't seen any reasons in the hobbit logs. Other RRD graphs are fine (disk, vmstat, etc) This is not just one night, but every night. I am using the hobbit client on the hobbit server, and hobbit client on another linux box (both are CentOS 4). Has anyone else seen this? What is significant about CPU, Proc/Users graphs? Thanks Craig Cook -- Systems Monitoring Consulting and Support Services http://www.cookitservices.com