similar to: extracting CPU usage from xentop

Hello, I noticed that the CPU% reported by xentop appears to be too high. With dom0 idle xentop is reporting between 14-17% CPU utilization while at the same time top in dom0 is reporting 1-4% CPU utilization. I see that the TODO for xentop includes a task to make CPU% more accurate but looking at the code for xentop and libxenstat I''m not sure how to go about this. Assuming the CPU

Hi All I wrote a perl script for extracting csv file from xentop batch mode (-b) . This script extracts data（time and cpu%) in CSV format. Please use it and comments are welcome! Thanks, Shao _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

This patch fixes the problem with xentop where it returns an exit code of 0 when passing it invalid options when it should return 1. The issue comes from getopt() returning ''?'' and the case from switch matching it. The solution was to remove the case ''?'' so that the default match could be triggered. Quote from `man 3 getopt` "If getopt() does not

# HG changeset patch # User inakoshi.hiroya@jp.fujitsu.com # Date 1191287395 -28800 # Node ID 5543e74774a826b1781893982ed5052312b820fc # Parent 83239b2890723e0c06bad507bb273a970784b18e Flush stdout when xentop -b gets SIGINT and SIGTERM. It is useful when you stop xentop -b by keyboard interrupt or by other programs such as killall from a batch script. You would have missed the bottom part of

# HG changeset patch # User john.levon@sun.com # Date 1172534164 28800 # Node ID 37fce3c0c10baee370ecc39b6a50fed8b232bd69 # Parent 4998793aa87729fbe8a582faa3a22c9ecfefe13c Fix for Solaris compile/output for VBDs in xentop. Signed-off-by: John Levon <john.levon@sun.com> diff --git a/tools/xenstat/xentop/xentop.c b/tools/xenstat/xentop/xentop.c --- a/tools/xenstat/xentop/xentop.c +++

This patch make xentop output statistics about both blkback and blktap. And, xentop also display back-end driver type in detail view mode. # HG changeset patch # User s-uchida@ap.jp.nec.com # Date 1180665749 -32400 # Node ID 1b36d6f6c7a98728382de93ee23c8c97b0b95dbd # Parent 1768a89bb87824249827d935d7803d8193bdeb05 XENTOP: Displaying blktap information. Signed-off-by: Satoshi UCHIDA

Hi, When I tested the xentop with a wrong option, the xentop returned the return code 0. # xentop --abc xentop: unrecognized option `--abc'' Usage: xentop [OPTION] Displays ongoing information about xen vm resources <snip> # echo $? 0 If the wrong option was specified, I think that the xentop had better return the return code 1 (or not 0). This patch fixes the return code of

Add periodic fflush to xentop(batch mode). If you want to make monitor tool using the following command, you can not get output every some seconds. xentop -b -d 1 > xentop.log (above tool use tail command to get periodic data from xentop.log) This patch is useful when you get xentop output(batch mode) periodicaly using pipe. Best Regards, Yusuke Kaneki ---- *** old/xentop.c

Hi. These patches add the output function for blktap statistics into xentop. Currently, xentop outputs only blkback statistics, and not blktap statistics. By these patches, xentop is made cover both blkback and blktap. In xen, back-end driver has two types; blkback and blktap. Blkback driver is old back-end driver. Blktap driver is new back-end driver and is recommended currently. Usage of both

Hello to all, Does anyone know if there is an iteration option planned for xentop (similar to top) to make it easier for simple shell scripts to scrape the output? I don''t see one in tools/xenstat/xentop/xentop.c , but I''ve been meaning to play with stuff in libxenstat anyway. If its going to be added soon I''ll tinker much later than sooner. TIA :) Best, -Tim

Dom0 is on Dell T310, Lenny with kernel 2.6.32.26 for jeremy git (commit b92d1ec0a86acd2dfe14d732c6a9304aae0ccf26) and xen 4.0.1 After double bridge with this config xentop not work and do segfault: ----------------------------------- /etc/xen/xend-config.sxp --- (network-script network-bridge-double) ----------------------------------- /etc/xen/scripts/network-bridge-double --- #!/bin/sh

Just downloaded the xen-unstable and updated it using mercurial. When i try to do a make install on a Celeron D 64bit processor, i get the following error: xentop.c:291: error: `KEY_DOWN'' undeclared (first use in this function) xentop.c:294: error: `KEY_UP'' undeclared (first use in this function) xentop.c:314: error: `KEY_BACKSPACE'' undeclared (first use in this

Hi All, I''m a beginner in Xen. Recently I''ve configured one VM (vm03) hosted in my physical machine. I ran the "netperf" workload in the vm01, and used the "top" command in the vm03 to test the behavior of the running workload. Meanwhile, I also used the "xentop" command in domain0 to see the resource consumption in vm03. Interestingly, I found that

Hi, Can someone please tell me if there is a multiprocessor version of xentop available or in the making ? I notice that the current xentop does not give any information per physical cpu on a multiprocessor machine. Thanks and regards Arjun _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users

Hi, I have a couple of questions. on Xentop: In Xentop what does the NETS column mean. I see a NETTX and a NETRX, and i think they mean the number of network bytes transmitted and received. Also what does the VBDs number mean. Is that the number of block device IO operations? or the number of IO bytes. on vcpu-list: I expected that at any point of time a pcpu would have only 1VMID in

I sent before few days. However, anyone did not comments. So, I resend this patch series. I hope to response for this patch series. This patch series provides the function to display VBD information through xentop. VBD information will be possible to get through sysfs which is locale in /sys/devices/xen-backed/vbd-x-xxxx, by previous patch. In this patch series, libxenstat collects these

I am running xen 3.2-1 (Debian 3.2.1-2) on Debian Lenny (2.6.26-2-xen-amd64 #1 SMP Wed Mar 10) on two identical production systems. This morning on one of the systems, xentop stopped displaying. Running xentop results in: unable to open xenstore : Connection refused Failed to initialize xenstat library xm list: Error: (9, ''Bad file descriptor'') xm dmseg: Error: Subcommand

Good suggestions - you are right that we only have issues when a VM is in transition so simply removing it from the returned list should be fine. Turns out that this also means the collectors don''t need to return special codes either - they are back to either working (and potentially pruning a domain from the list) or failing completely. Patch attached (boy, this is taking waaay more

I am creating some custom nagios plugins and I am scraping data from xen using the nagios command: /usr/sbin/xentop -b -i 2 -d1 I am trying to get the network tx / rx data, however I think I am missing something here. NETTX(k) NETRX(k) So total network bandwidth is 1048576 k if the above header from xentop is true. So I am wondering why the output for each vm exceeds the theoretical max, Is

Hello, I found the result of "xentop" differented with the result of "top", I think this may be the "top" can not monitor the load of xen hypervisor, how can I get the load of xen hypervisor, thanks! 2008-10-17 xdongp _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users