Re: RX port buffers on cat4000's [7:61248]
* In-Lost - Packets which could not be received since the input buffers are full. Reason: Excessive input rate of traffic. * Delay Exceed - This is an indication of the number of frames discarded because of excessive delay in the switching process. Reason/Cause: Severe problem with the switch. Open a case with the Cisco TAC I'm guessing the PIX connection has lots of traffic (probably constant). maybe some sort of attack was going on at this time. Might be a combination of devices attached to that blade. There is no buffer adjustments I know of. Also the sup2 on 4006 does 18 Mpps , whereas a sup3/sup4 can do 48 Mpps. --- "Vicuna, Mark" wrote: > Hi All, > > Just wondering if anyone has figures for the size of > rx buffers for cat 4000 > ports? Had a issue today where a port was connected > to a pix 535 manually > set at half/100 (yep you read right), the switch > port was at auto/auto. The > rate of In-Lost (rx buffer filling up) errors was on > average 5 per minute > (among all the other errors of course). > > I have seen In-Lost and delay-exceeds rise up for > mis-settings to servers, > but the pix connection was showing some pretty fast > counter stats > > Hard to find these small details sometimes in doco.. > maybe anyone here from > cisco can advise? > > the mod on the 4006 is a ws-x4424-gb-rj45 (hw 1.5) > with a supII (hw 3.2, > gsp 7.1(2.0), nmp 7.1(2)) > > > Cheers, > M [EMAIL PROTECTED] __ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61251&t=61248 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: RX port buffers on cat4000's [7:61248]
yes, alot of traffic is flowing through :-) and no wasn't related to any kind of attack. The 18mpps or 48mpps is more to do with the switch fabric, not related to the tx rx buffer(s) that are allocated per port. the issue was immediately resolved when i hard coded port settings. just wanted to have an idea of the size of buffering allocated on these ports (hardware specific of course). cheers, mark. -Original Message- From: Erick B. [mailto:[EMAIL PROTECTED]] Sent: Thursday, January 16, 2003 11:17 PM To: Vicuna, Mark; [EMAIL PROTECTED] Subject: Re: RX port buffers on cat4000's [7:61248] http://www.cisco.com/warp/public/473/46.html * In-Lost - Packets which could not be received since the input buffers are full. Reason: Excessive input rate of traffic. * Delay Exceed - This is an indication of the number of frames discarded because of excessive delay in the switching process. Reason/Cause: Severe problem with the switch. Open a case with the Cisco TAC I'm guessing the PIX connection has lots of traffic (probably constant). maybe some sort of attack was going on at this time. Might be a combination of devices attached to that blade. There is no buffer adjustments I know of. Also the sup2 on 4006 does 18 Mpps , whereas a sup3/sup4 can do 48 Mpps. --- "Vicuna, Mark" wrote: > Hi All, > > Just wondering if anyone has figures for the size of > rx buffers for cat 4000 > ports? Had a issue today where a port was connected > to a pix 535 manually > set at half/100 (yep you read right), the switch > port was at auto/auto. The > rate of In-Lost (rx buffer filling up) errors was on > average 5 per minute > (among all the other errors of course). > > I have seen In-Lost and delay-exceeds rise up for > mis-settings to servers, > but the pix connection was showing some pretty fast > counter stats > > Hard to find these small details sometimes in doco.. > maybe anyone here from > cisco can advise? > > the mod on the 4006 is a ws-x4424-gb-rj45 (hw 1.5) > with a supII (hw 3.2, > gsp 7.1(2.0), nmp 7.1(2)) > > > Cheers, > M [EMAIL PROTECTED] __ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61253&t=61248 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: RX port buffers on cat4000's [7:61248]
all ports share a 16MB buffer space. (4000 docs) for ethernet interfaces, i only found a generic allocation of ether and fast ethr blades, 192KB per interface, 160 for outboud and 24 for inbound, but i found this info in the (5000 docs), they made it sound like it's generic to ethernet but i'm not sure. ""Vicuna, Mark"" a icrit dans le message de news: [EMAIL PROTECTED] > yes, alot of traffic is flowing through :-) and no wasn't related to any > kind of attack. > > The 18mpps or 48mpps is more to do with the switch fabric, not related to > the tx rx buffer(s) that are allocated per port. > > the issue was immediately resolved when i hard coded port settings. just > wanted to have an idea of the size of buffering allocated on these ports > (hardware specific of course). > > > cheers, > mark. > > -Original Message- > From: Erick B. [mailto:[EMAIL PROTECTED]] > Sent: Thursday, January 16, 2003 11:17 PM > To: Vicuna, Mark; [EMAIL PROTECTED] > Subject: Re: RX port buffers on cat4000's [7:61248] > > > > http://www.cisco.com/warp/public/473/46.html > > * In-Lost - Packets which could not be received since > the input buffers are full. Reason: Excessive input > rate of traffic. > > * Delay Exceed - This is an indication of the number > of frames discarded because of excessive delay in the > switching process. Reason/Cause: Severe problem with > the switch. Open a case with the Cisco TAC > > I'm guessing the PIX connection has lots of traffic > (probably constant). maybe some sort of attack was > going on at this time. Might be a combination of > devices attached to that blade. There is no buffer > adjustments I know of. > > Also the sup2 on 4006 does 18 Mpps , whereas a > sup3/sup4 can do 48 Mpps. > > --- "Vicuna, Mark" wrote: > > Hi All, > > > > Just wondering if anyone has figures for the size of > > rx buffers for cat 4000 > > ports? Had a issue today where a port was connected > > to a pix 535 manually > > set at half/100 (yep you read right), the switch > > port was at auto/auto. The > > rate of In-Lost (rx buffer filling up) errors was on > > average 5 per minute > > (among all the other errors of course). > > > > I have seen In-Lost and delay-exceeds rise up for > > mis-settings to servers, > > but the pix connection was showing some pretty fast > > counter stats > > > > Hard to find these small details sometimes in doco.. > > maybe anyone here from > > cisco can advise? > > > > the mod on the 4006 is a ws-x4424-gb-rj45 (hw 1.5) > > with a supII (hw 3.2, > > gsp 7.1(2.0), nmp 7.1(2)) > > > > > > Cheers, > > M > [EMAIL PROTECTED] > > > __ > Do you Yahoo!? > Yahoo! Mail Plus - Powerful. Affordable. Sign up now. > http://mailplus.yahoo.com Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61297&t=61248 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
