Hide Forgot
Description of problem: When the irqbalance service is started within our new Intel SandyBridge-based servers with an amd64 kernel, the system hard locks up within 30 seconds. Version-Release number of selected component (if applicable): irqbalance-0.55-29.el6.x86_64.rpm How reproducible: Steps to Reproduce: 1. Boot system with 64-bit kernel. 2. Start Redhat-provided irqbalance service. Actual results: The system locks up within 30 seconds. "Locked up" means that the system no longer accepts keyboard input, the system drops all network responsiveness and the server is no longer accessible until the hardware of the system is reset. This error message appears to print continually while the system is locked up: No irq handler for vector (irq -1) Expected results: Working system that continues to stay stable. Additional info: This issue can be resolved immediately by retrieving the latest version of irqbalance from http://irqbalance.org/download.html (irqbalance-0.56.tar.bz2) compiling the software in the same server and supplanting the Redhat provided irqbalance binary with the newly compiled binary. Please let me know if any hardware information is needed. I have the systems running and can provide any information necessary. Thanks in advance for your assistance.
This request was evaluated by Red Hat Product Management for inclusion in the current release of Red Hat Enterprise Linux. Because the affected component is not scheduled to be updated in the current release, Red Hat is unfortunately unable to address this request at this time. Red Hat invites you to ask your support representative to propose this request, if appropriate and relevant, in the next release of Red Hat Enterprise Linux. If you would like it considered as an exception in the current release, please ask your support representative.
Hi Davy, could you please specify your HW? I've been not able to reproduce it. Outputs of 'dmidecode' and 'lspci -v' would be sufficient. Thanks, Petr H
CLOSED as INSUFFICIENT_DATA. If the problem would be still here, feel free to reopen the bug.