Cisco WAN :: 2851 And 1841 / Buffer Failures On Boot?
Oct 17, 2012
I have a 2851 and a 1841 both serving as hub routers in a GRE multipoint configuration. They are both receiving buffer misses and failures on startup. I will post the output of show buffers below:
LAB-HUB-RTR#show buffers
Buffer elements:
607 in free list (500 max allowed)
9071 hits, 0 misses, 618 created
Public buffer pools:
Small buffers, 104 bytes (total 71, permanent 50, peak 71 @ 00:11:33):
68 in free list (20 min, 150 max allowed)
7083 hits, 74 misses, 0 trims, 21 created
55 failures (0 no memory)
[code]....
I have tried increasing the small/medium buffers initial size and permanent size, however there is no change. The buffer failures for small and medium buffers are always around this many every boot.I have also changed the IOS versions between 12.4.24(T4) to 15.1 with no luck in stopping the failures.
I recently obtained a 2851 and have been battling this issue for a week now. An odd set of circumstances happen with the bootstrap startup process. If the router has loss of power or when reloaded; it fails to read the CF card and boots to rommon. I can tftp an image to it (tftpdnld -r) and IOS CAN read the flash card along with any saved configuration.In fact once in IOS, I can wr mem and print the hardware details of the CF card. Once I reload the router it goes to rommon, fails to open the flash, then sits until i tftp the image back. When the router load IOS, it also loads the startup config.I don't think it's a problem with the CF card as IOS can read it. I've tried formatting the card from IOS and from my pc as FAT, FAT16, FAT32, and NTFS - it didn't make a different. I updated the ROMMON to allow for USB booting but haven't purchased a USB Drive yet for testing. The upgrade didn't resolve this issue either. It almost seems like a firmware problem with the EPROM.
we have a problem with a Cisco 1841 router while booting. The router is loading the image from the CF card but gets stucked at this point:
This product contains cryptographic features and is subject to UnitedStates and local country laws governing import, export, transfer anduse. Delivery of Cisco cryptographic products does not implythird-party authority to import, export, distribute or use encryption.Importers, exporters, distributors and users are responsible forcompliance with U.S. and local country laws. By using this product youagree to comply with applicable laws and regulations. If you are unableto comply with U.S. and local laws, return this product immediately.
A summary of U.S. laws governing Cisco cryptographic products may be foundURL
Getting message, cannot open "flash:" on 1841 evdo router, how do I reinstall flash? I am currently consoled into the router, the prompt is coming up as rommon 1>
I have a cisco1841 router with 32M CF card installed.Now , i want boot a system image larger 32M from usbdisk.First i insert the 128M usb flash disk to the 1841 router and format it, then i can see the device from dir command.Second copy the larger ios image from tftp to the usb disk and successed,i can see the image with " dir usbflash0:" command.Third i use this command "boot system " ,but there is no selection of "usbflash0: " Does the cisco 1841 router boot the system image from usb disk when reboot the router.
My netbook keeps getting DNS Lookup Failure messages. This happens about every 20 minutes, give or take, though sometimes it can go longer before the failure. I can "solve" the problem by quickly repairing the network connection, but that's tedious. It happens only on my laptop and not on my housemate's desktop (wired connection) though he did say that it happens to him on his laptop as well. In looking around I thought the problem might be because i'm using Chrome, but it happens on Firefox and Opera as well.
I have over 20 units doing the same thing and it seems to be a software isssue but i dont see any bugs or posts on it. This is only on 2960S switches and not 2960 or 2960G units.
If i use the password reset feature to break the units into rom and then type "boot" instead of power cycling the unit, they will fail MBIST post tests. If the unit is power cycled or left to boot normally on its own, there is no issues and all post tests pass. I know MBIST is Memory Built In Self Test and was thinking maybe breaking the unit into rom disrupts those memory tests for some reason. I tried the following software and got the same results with all of the images:
122-55.SE2 122-55.SE 122-53.SE2 122-53.SE1
Logs attached are from the same switch, one with password reset procedure used and while left to boot on its own.
My internet connect works fine for hours at a time, then suddenly will get 7 or 8 General Failures while pinging (long enough to boot me off the net) then will be back to working fine immediately afterwards.Pinging 127.0.0.1 works fine. I've checked the firewall (Norton) and it has the default settings
C3745-ENTSERVICESK9-M version 12.4(10b). ROM version 12.2(8r)T2.This router appears to be generating spurious fan failure alarms. The fan assembly has been changed for a new one, the engineer checked and was satisfied that in a physical functional sense the replacement fan assembly was definitely working, but we now have all 4 fans showing as failed. We are being told that the replacement unit is believed to be part of a batch of faulty fan assemblies and that it was a known Cisco issue.
I have a site to site ipsec tunnel setup between an ASA5510 and a 2951 Router. The ASA 5510 is on a 10.x subnet with a few vlans behind it. There are also 7 other ASA5505 that connect to this box with ipsec.
The 2951 is on a 10.x subnet with multiple vlans behind it (10.x and 192.x subnets).
When I had ACL to allow traffic from 10.20.0.0 (ASA) to 192.168.111.0 (2951 - voice vlan) the connection comes online and is stable.
The minute I add any of the following, the connection drops off with Phase 2 errors: 10.20.0.0 to 10.1.200.0 10.20.1.0 to 10.1.1.0
I can add a second 10.20.0.0 to 192.168.10.0 with no problem at all. The issue only seems to occur when attempting to add traffic from 10 to 10 on the tunnel.
I manage a VPN 300 concentrator which has been happily working for several years without any problems. All users are part of the same group and authenticate to an RSA server. We recently moved from RSA authentication manager 6.1 to RSA authentication manager 7.1. Everthing continued working fine for several weeks, then at the beginning of this week we started getting users intermittently failing to connect to the VPN. I'm not sure if this problem relates to our new RSA server, but we have other network devices which authenticate to it with no problem so I guess the problem is with the VPN concentrator itself.
When users fail they just get a generic "Reason 427 connection terminated by peer" error message. The live event log shows "group = vpn, status = Not-in-service" when their connection fails. Other times they connect normally and no error messages are displayed. There seems to be no real pattern, sometimes your connection fails but if you keep trying you will eventually get in [however it can take many attempts over an hour or two before you succeed, or you may get in straight away with no problem].
I dont believe its a network problem, as I have run continuous pings to the concentrator and the RSA server whilst users are experiencing these problems and there are no drops.
The RSA servers authentication monitor always shows that the user has successfully authenticated, whether the users connection actually succeeds or not. I am tempted to just reboot the concentrator, but we have site-to-site VPN tunnels connected off it and I'm a little concerned that if it is faulty it may not come back up at all.
I'm looking for a way to monitor client authentication failures with our 3 standalone 1142N APs. I know that I can see failures under the log viewer of each AP
After upgrade from ACE20 with A2(3.5) to ACE30 with A5(1.2) I get failures in a number of server farm's, where before upgrade the number was zero. No drops in VIP and logs from applications do not notice any new errors.
I made a custom-built V1 Windows Home Server that I really would like to be able to remote access. I have tried the Netgear 3700, but it did not allow remote access. A D-Link DIR-825 does, but it, and many D-Link products, have a persistent problem of requiring a reset due to dropped connections. I have had 2 of the DIR-825's drop connections. I have been told that their QoS components cannot handle the load on them and fail, causing the drops, but I cannot corroborate that.Perhaps what I need is a router that allows "NAT loopback"? This way I can see the WHS Console verify that I can access the server from outside my network. I have tried to do so with the above routers via a 3G connection on my iPhone 4 and all except the D-Link failed to allow access to my WHS.
I should add that I am using a D-Link DSL-520B modem on ATT DSL. It is a 6MB connection from the ISP. Previously the modem was in "bridge mode" on the D-Link router. Also, contacting ATT I was told they do not block any ports. I have tried forwarding the proper ports (80, 443, 4125) for the WHS, but that has not given me remote access. I did get them by enabling UPnP on the D-Link. Is all this an issue of needing the modem on "bridge mode" in order to work properly? Any router for my needs that allows remote access (NAT loopback needed?) and also has a solid connection? Gigabit ethernet is a must have too. Otherwise I am open to options. I would like a combined router/modem unit to make things a little easier.
When mls qos is disabled, how much of an interface buffer is available to the single output queue?is a single switch port able to consume as much as required of the common memory pool.
I would like to test a wireless connection by streaming video from one device to another,I would like to monitor the size of a playout buffer to gauge whether or not the link can sustain a high quality HD stream. Any software that will allow me to monitor the playout buffer?
I'am a bit newbie at using Cisco products and here is my problem : I have set up a VPN tunnel between 2 Sites (A and B) a few month ago using 2 cisco SR520-ADSL-K9. All was working fine until power failures occured on the sites B (secondary site).
What happened was that none of the ethernet ports were working, excepting during booting, I was then able to ping computers linked to ports Fastethernet0, FastEthernet1, FastEthernet2 and FastEthernet3 but after a few seconds all ports were disabled but my DSL seemed to be working.
So I took back the router home to check it. I managed (I think) to make a factory reset using a serial terminal and following the procedure described here [URL]
Since I did the reset, I thought I would be able to re-use Cisco Configuration Assistant (3.1) to re-configure the router (I am very bad at using the command lines) but I am unable to connect to the router using the supposed default IP : 92.168.75.1 (I set my computer to use 192.168.75.50 IP adress with mask 255.255.255.0). But I can't connect to the router ... even if the Ethernet ports seem to work because green light is on when plugging my cable. connect to my router using CCA ?
For more information, here is what I get when I run "show startup-config" and "show running-config" in terminal console. I guess the objective is to make the startup-config beeing the running-config, but I have no idea on how to do that ..
show startup-config show running-config Router#show startup-config
I have seen this at two sites now: after migrating the site T1 to 10-Mbps Opt-E-MAN and replacing old 10/100 switches with 10/100/1000 switches, users frequently get http connection errors. The error goes away if the user reloads the page--sometimes they have to reload more than once. They never had this problem before.I thought it was due to the large number of 5-port desktop switches infesting the networks (I'm getting rid of them as fast as I can) but it happens even on a PC directly connected to one of the new GigE switches. It does not happen when accessing internal web pages. It looks like a DNS failure -- but nothing has changed in our DNS setup, except that users have a fatter pipe to our DNS servers.
Installed Cisco AnyConnect Secure Mobile Client on a new Asus CM6870, downgraded to Windows 7 Pro. It worked fine for 3 days, establishing VPN connections with my workplace without a probllem. Then it repeatedly failed to connect.
I attempted an uninstall/re-install, and the install now fails as well, returning the following error: The VPN client agent was unable to create the interprocess communication depot. When I do manage to get it re-installled, it works sometimes, then fails to establish connections other times. I am not an IT professional, so trying to diagnose the issue by reviewing the Windows/Inf/setupapi.app.log and .dev files is a no go. I do not hold a contract with Cisco so I am not authorized to open a support ticket, or receive phone support (again, I tired).
In each case, the routers have functioned flawlessly for a period of 2 to 8 months, then suddenly begin to require daily to hourly reboots to keep the speed up, and often times fail to respond to any web activity whatsoever. Ping tests are intermittent, sometimes failing but other times succeeding while web sites remain unresponsiveFor three years I lived with two or more room mates at a time, each of us with our own computer (or 2) and all doing a lot of peer to peer downloading. I realize a router can overheat during heavy use like this, so buying new routers so regularly has seemed vaguely understandable. However it doesn't seem like that is the case in my most recent failure.I've lived alone for the past 3 months, and have owned the Belkin Play Max N600 HD since. I have NOT been P2P downloading or putting a heavy load on my router in any way (or so it seems). Yet as of about two weeks ago, it has suffered a major slowdown just like all of its predecessors. Yet my 30 mbps internet connection roars to life the moment I plug the modem directly into a computer.
My desktop remains on 24/7 but like I said before, I do not do constantdownloading/uploading. Both wired and wireless connections are effected equally, and I have always kept all my routers WPA encrypted.When websites become unresponsive on my Belkin today, it is usually after everything has been sitting idle for some time (overnight, or all day while I am at work). Speaking of work, today is a perfect exampleI remoted into my home desktop and was able to interact just fine, yet when I would launch a browser and try to load any website at all, I get absolutely nothing. I had to transfer a text document through DropBox (which also still worked) because I couldn't get Google Docs (or gmail) to load on the remote computer
For more than 6 months I have been happily using a Sierra Wireless AirCard313U to connect my Win7 Pro (64-bit x86) laptop to the internet, and using another corporation's VPN (Citrix Access Gateway) as a contractor - so networking problems ARE my problem.
Now what happens is: DNS failures halt my internet browsing and any hope of RDP'ing into the corporate LAN whenever I connect to the VPN while using the AirCard (its a USB/cell modem device). The DNS lookups and RDP'ing works fine when I'm using the VPN with my wifi.
What changed recently:Installed Connectify software Uninstalled Connectify software (didn't work w/AirCard) Installed VirtualRouter software Uninstalled VirtualRouter software (didn't work w/AirCard) Installed Sierra Wireless' AirCard Watcher software (manager app, may have included updated drivers for the AirCard) Re-install Connectify software (now worked w/AirCard)Uninstalled Connectify software (DNS problem had appeared)
I've tried removing ISATAPs, running these commands & rebooting, to no avail:netsh int ipv4 reset reset.log netsh int ipv6 reset reset.log netsh winsock reset catalog
Right now, connected to VPN via wifi my ipconfig /all looks like this:
Windows IP Configuration Host Name . . . . . . . . . . . . : W7 Primary Dns Suffix . . . . . . . :
I use to could be able to watch videos and listen to music, and download at optimal speeds, I could even watch HD if I wanted to, but now all of a sudden (starting at least 3 weeks ago) videos, downloads, and music seem to run super slow no matter what quality I have it on. It takes 5 minutes to buffer a 3 min video, and if I want to watch a show, it take like 2 or 3 hours to fully buffer. My internet still goes fast, and I have trendmicro antivirus, as well as ccleaner, i updated my drivers, i have ie9 and adobe flashhplayer 11.1, i updated everything in windows update, i unplugged the modem for awhile and shut down more often, and opened it up and blew all the dust out. I don't know what else to do. I do have a cisco E2000 wireless router if that's a problem, and both my ps3 and my computer are on the connection, but my computer was working fine with both on the same connection for a year. I go on the http:192.168.1.1 cisco place, and i renewed ip settings, and nothing worked.
About a week ago, without any change to my network settings, I began having a problem with streaming 1080p content from my media server to a router, wired to a switch that serves a few devices in my home theater (including a HTPC, my Boxee Box). Every 15 seconds movies have to buffer. After a while, I narrowed it down to the switch, since wiring the Boxee Box straight to the router caused the movies to run perfectly.Logically, I thought the switch was the problem, since I really made no changes on my network/DD-WRTRouter other than running a Minecraft server (which shutting that down made no difference in network performance).
I ordered a brand new gigabit switch, same problem. Pinging the BoxeeBox returns 0 packet loss whether it's connected to the switch or router, but movies still play perfectly fine when directly connected to the router. Different cables didn't work. Ok, well, I thought maybe the other devices on that switch were causing some sort of conflict.. changed the BoxeeBox's static IP just in-case, unplugged every other device from the switch so that only the BoxeeBox and Router were connected to the switch. It worked fine! Yay! Tried a different cable, still worked fine. Great, I thought, so the problem was never my switch... plugged the old switch back in, then the problem returned.
Changed back to my new switch (each time power cycling all of the network devices) and the problem persisted, even with the SAME EXACT setup I just had a few minutes previously that worked fine. Tearing my hair out at this point.Since that moment I still haven't been able to reproduce a perfect stream unless the BoxeeBox is directly connected to the router. This doesn't make ANY sense as I had this setup running perfectly fine since November 2010. I am COMPLETELY out of ideas other than running multiple ethernet wires to that drop or running all the other devices on wireless.
I would like to define a general policy for our company to set the log buffer ("logging buffered xxx"). Currently, most of our configs use the default value of 4096, which does not store much of the history. I'd like to see more, when logged onto a router. We are using different router platforms (from old 1700 up to 39xx oder 72xx) Can you give me any hints, how large the buffer can be set on a platform? Which checks should I do to determine, how large I can set the buffer?
Cisco says: "Total = the total amount of memory available after the system image loads and builds its data structures."So, in this example, the total bytes are 17803244, which is ~ 16.9 Mbyte. So, if I allocate 64000 bytes, my total buffer will decrease to ~16,3 Mbyte.
I my calculation above is correct, I see no problem in defining a standard value of "logging buffered 64000" for every router platform - even under high load conditions. Since 64 kbytes is a very small value compared to the total usable memory of a Cisco 1720, newer platforms should not have any problems either.
I leave on one of my computers on all the time, an Acer desktop with Windows XP. I also have two laptops. I have Verizon DSL and a linksy wireless G router. When we have a power outage the Acer desk top loses it's internet connection. Trying to reconnect the connection takes hours. Even when we finally get it back we have no idea how we did it. The laptops which of course were off have no problem.
I just built a new computer running Windows 7 with an Intel 82579V Gigabit on the motherboard. Since last night while playing an online game, I have occasionally noticed sudden network failures where I either time out of my game or severely lag for 10 seconds before the game catches up.
My other network hardware and setup remains unchanged during this build.I have never had issues with my internet connectivity from my ISP.
The problem is not consistent and I'm not sure how to repeat it. I was playing my online game for several hours and then suddenly started having problems every 5 minutes with disconnecting.I can watch the network utilization in the Windows Task manager drop suddenly and then spike back when I reconnect to the game. Similar visual seen on the Tomato software on my router.
So far I've grabbed the latest drivers for the network chip from my motherboard manufacturer's website and turned off Teredo. I also made sure no power saving features were enabled on the network chip in the device manager.
Few days ago I faced a issue in which one of ours 4500 stopped providing poe on some ports in one line card. I called TAC and ran some diagnostics. However, there is a command that I found and it is NOT DOCUMENTED on 4500 reference guide! The command is "diagnostic monitor poe". This command actually detects/recover POE hardware failures! After executing this command, poe start to work again?
We're having some issues with a 3560 CPE. It's uplink is a GE fiber link, customer port is FE RJ45. We see a lot of TX-frames being dropped at the FE port,but none at the GE port. Even when the customer is only at ~50Mbit/s downstream traffic. When customer sending ~50Mbit/s upstream, there are no TX- drops at the GE link. Is this a normal behaviour? From what I know the physical medium shouldn't have any inpact on this since drops occurs in the port- asic, and not in physical transmission.Do the buffer-sizes between GE and FE differ? What could we do to optimize the flow and reduce drops? QoS is set to off and no modifications on the queues have been made on the interfaces.
