When you were "firing" the box before production - was the ethernet traffic generated from disk usage, or just packet tossing? Two of the boxes I had for an older setup would behave as long as there wasn't a lot of interrupts being generated, but the moment I started a lot of disk usage + iscsi + nfs, the nics I was using would start dropping packets fairly regularly. It usually isn't too big of a deal to change out nics, but in this instance it was easier to just disable the broadcom offload functions and let linux handle everything.
-----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Evgeny Yurchenko Sent: Tuesday, September 30, 2008 7:23 AM To: [email protected] Subject: RE: [OF-users] Re: Re: Error message The NIC is loaded only about 250Mbit/s during when it happens. I fired this box before I put it in production up to 700-800Mbit/s (I mean Ethernet traffic). Now we put a lot of statistics to log in place. Waiting for the problem to reoccur. We’ll inform you later. Thanks. Eugene ________________________________ From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jason Litka Sent: September 30, 2008 10:21 AM To: [email protected] Subject: RE: [OF-users] Re: Re: Error message Broadcom NICs are known to flake out under Linux when experiencing high load. Try switching to an Intel Server NIC. Jason Litka | Xoxide, Inc. Manager, IT & Software Development P: (610) 251-1672 x200 F: (610) 251-0263 From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Roman Naumenko Sent: Tuesday, September 30, 2008 8:54 AM To: [email protected] Subject: [OF-users] Re: Re: Error message On Tue, 2008-09-30 at 07:00 -0400, [EMAIL PROTECTED] wrote: ---------------------------------------------------------------------- Message: 1 Date: Mon, 29 Sep 2008 17:17:33 -0400 From: Roman Naumenko <[EMAIL PROTECTED]> Subject: [OF-users] Error message To: [email protected] Message-ID: <[EMAIL PROTECTED]> Content-Type: text/plain; charset="us-ascii" We have error in /var/log/messages on one of our most loaded SAN servers: kernel: iscsi_trgt: Logical Unit Reset (05) issued on tid:1 lun:0 by sid:1971425271414848 (Function Complete) As we discovered this error appears when Exchange heavily writes its logs on target device. We are still investigating disk loads on the server trying to figure out if error are actually related to disk load. rPath Linux 2.3 2.6.24.7-0.5.1.smp.gcc3.4.x86.i686 >free: total used free shared buffers cached Mem: 2075368 536492 1538876 0 65964 324060 -/+ buffers/cache: 146468 1928900 Swap: 2097144 0 2097144 What else should we look into? Regards, Roman Naumenko -------------- next part -------------- An HTML attachment was scrubbed... URL: https://lists.openfiler.com/pipermail/openfiler-users/attachments/20080929/4c7de0b6/attachment-0001.html ------------------------------ Message: 2 Date: Mon, 29 Sep 2008 17:20:13 -0400 From: "Jason Litka" <[EMAIL PROTECTED]> Subject: RE: [OF-users] Error message To: <[email protected]> Message-ID: <[EMAIL PROTECTED]> Content-Type: text/plain; charset="utf-8" What kind of network cards are you using? >lspci -v 03:00.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5721 Gigabit Ethernet PCI Express (rev 21) Subsystem: ASUSTeK Computer Inc.: Unknown device 8149 Flags: bus master, fast devsel, latency 0, IRQ 220 Memory at ff7f0000 (64-bit, non-prefetchable) [size=64K] Capabilities: [48] Power Management version 2 Capabilities: [50] Vital Product Data Capabilities: [58] Message Signalled Interrupts: 64bit+ Queue=0/3 Enable+ Capabilities: [d0] Express Endpoint IRQ 0 We have other SANs servers with such cards, they work fine. Regards, Roman Naumenko _______________________________________________ Openfiler-users mailing list [email protected] https://lists.openfiler.com/mailman/listinfo/openfiler-users
