Date: Sun, 06 Jan 2013 09:39:41 +0900 (JST) From: moto kawasaki <moto@kawasaki3.org> To: freebsd-xen@freebsd.org Subject: XENHVM hang-up w/ "network_alloc_rx_buffers: m_cljget failed" Message-ID: <20130106.093941.323021822222343593.moto@kawasaki3.org>
next in thread | raw e-mail | index | archive | help
Hello,
Thank you very much for your effort to porting Xen to FreeBSD.
I am so happy I can use FreeBSD instead of Linux on XenServer!!
Today, I have encountered hang-ups as follows;
- XenServer 6.1 * 3 hosts (HP DL360G7)
- FreeBSD 9.0-RELEASE-p5 (customized XENHVM) * 2 VMs
- Those VMs hang up with the following messages on console:
network_alloc_rx_buffers: m_cjlget failed
This message is repeatedly shown so quickly, and I cannot get
login prompt.
- During this issue, I cannot login via ssh over network, although
TCP connection is established (I tried with telnet to ssh port).
- I guess the trigger of this issue was something like broadcast
storm, which was made by failed network configuration change.
- This issue seems similar to exhausting mbuf(rx).
Could you please advice me of countermeasure/prevention against this
issue ?
Also, I would like to know how to make it self-healing, I mean I want
to make those VMs to recover automatically after the trigger had gone.
Any suggestion/hint/reference is welcomed.
Thank you very much.
--
moto kawasaki <moto@kawasaki3.org>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20130106.093941.323021822222343593.moto>