From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.8.0.8) Gecko/20061108 Fedora/1.5.0.8-1.fc5 Firefox/1.5.0.8 Description of problem: With increasing load, nfsd causes a kernel panic and system becomes unresponsive. Problem occurs with number of clients above 30. System generally stable with lower numbers or relatively inactive clients. Heavy use in automating testing environment (5 machines constantly accessing test files) will cause kernel panic in roughly 24 hours. Version-Release number of selected component (if applicable): 1.0.10-5.fc6 How reproducible: Always Steps to Reproduce: 1. Export a files system via nfs 2. Mount from several machines (more machines will cause more frequent failure) 3. Continually access files (reading only) 4. Kernel panic will occur Actual Results: Expected Results: Additional info:
Created attachment 144730 [details] Log of kernel panic error messages THis is the log from the machine which experienced the kernel panic
Do you have some type of disk mirroring set up
Yes. Nvidia SATA mirroring 2 250GB hard drives. Output of lspci: 00:0e.0 RAID bus controller: nVidia Corporation MCP51 Serial ATA Controller (rev a1) 00:0f.0 RAID bus controller: nVidia Corporation MCP51 Serial ATA Controller (rev a1)
(In reply to comment #2) > Do you have some type of disk mirroring set up Yes. Nvidia SATA mirroring 2 250GB hard drives. Output of lspci: 00:0e.0 RAID bus controller: nVidia Corporation MCP51 Serial ATA Controller (rev a1) 00:0f.0 RAID bus controller: nVidia Corporation MCP51 Serial ATA Controller (rev a1)
Further testing shows that if you add the kernel options maxcpus=1 nosmp . Machine is completely stable been running for 1-1/2 weeks with heavy load.
reassigning this back to kernel-maint since I really don't think this is an NFS bug...
What kernel version are you using?
(In reply to comment #8) > What kernel version are you using? > During original problem kernel version 2.6.18-1.2868.fc6 had to be booted with maxcpus=1 kernel option. Kernel version 2.6.19-1.2911.fc6 #1 SMP has been running 3 weeks without maxcpus option set and is stable.