Common Errors on the Linux Server front-end and their meanings

Article Title: prompts and meanings of Common Errors on the Linux Server front-end. Linux is a technology channel of the IT lab in China. Includes basic categories such as desktop applications, Linux system management, kernel research, embedded systems, and open source.

Common Errors on the Linux Server front-end:

◆ General class prompt eth1: Too much
　　
Work at interrupt, IntrStatus = 0x0001

The meaning of this prompt is. A network card has too many interrupted requests. if it appears only once, it can be ignored. however, if this prompt appears frequently or appears in a centralized manner, there are many possibilities to be addressed .. There are many possibilities, such as NIC performance, server performance, and network attacks.
　　
◆ General class prompts
　　
　　IPVS: incoming ICMP: failed checksum from 61.172.0.X!
　
The server receives an ICMP packet with a checksum and an error. This type of data packet may be generated illegally. However, the server has received a large amount of such data, which is generally ignored.

Generally, the proxy server forwards thousands of data packets every second during work. It does not affect normal operation if several error data packets are received. This is the most common type of prompt.

◆ General class prompts

　　NET: N messages suppressed.
　
The server ignores N Packets, which is similar to the previous prompt. Packets received by the server are considered useless junk data. This type of data is mostly produced by attack programs.

This prompt can be ignored if N is small. However, if such a prompt appears frequently or for a long period of time, it is likely that the server is under bandwidth attacks such as junk data.

◆ General class prompts

　　UDP: bad checksum. From 221.200.X.X: 50279 to 218.62.X.X: 1155 ulen 24

　　UDP: short packet: 218.2.X.X: 3072 3640/217 to 222.16x.x: 57596

　　218.26.131.X sent an invalid ICMP type 3, code 13 error to a broadcast: 0.1.0.4 on eth0

The server receives an incorrect packet, namely UDP checksum and error, too short UDP packet, and an incorrect ICMP data packet. This type of information is generally generated illegally. However, if the problem is not serious, you can ignore it directly.

◆ General class prompts

　　Kernel: conntrack_ftp: partial 227 2205426703 + 13

　　FTP_NAT: partial packet 2635716056/20 in 2635716048/2635716075

An error occurs when the server maintains an FTP connection. Such prompts can generally be ignored directly. Serious network communication problems!

　　

NETDEV WATCHDOG: eth1: transmit timed out eth1: link down eth1: link up, 10Mbps, half-duplex, lpa 0x0000 eth2: link up, 100Mbps, full-duplex, lpa 0x41E1 setting full-duplex based on MII #24 link partner capability of 45e1

These prompts appear only when serious problems occur in network communication. Faults are basically related to network disconnection. These prompts indicate that data transmission times out for a certain network adapter, the network adapter is down, the network adapter is up, and the connection rate is 10/100 Mbps. The prompts for the last three lines written here are similar. When this type of prompt appears, you must pay attention to the network connection status for handling !!!

◆ Serious network communication problems!

NIC Link is Up 100 Mbps Full Duplex

The connection speed is the same as that of kernel: eth1: link up. It is generally considered that the network adapter is down, but the adaptive speed of the network adapter is also a communication problem. This type of information can be ignored if the network cable is disconnected normally.

◆ Serious network communication problems!

Eth0: Transmit timed out, status 0000, PHY status 786d, resetting... eth0: Reset not complete yet. Trying harder.

The first prompt indicates that the NIC fails to send data. Reset the NIC. The second prompt indicates that the NIC is not successfully reset... These prompts are all serious communication problems.

◆Alarm program prompt

0001 ## WMPCheckV001 ## 2005-04-13_10: 10: 01 Found. (ARP Spoofing sniffer )! IP: 183 MAC: 5 0002 # WMPCheckV001 #2005-04-07_01: 53: 32 Found. (MAC_incomplete )! IP: 173 mac_incomplete: 186 0003 # WMPCheckV001 #: 25: 11 Found. (HIGH_synsent )! Totl: 4271 SynSent: 3490 0004 # WMPCheckV001 #20 ......

This is a prompt caused by the alarm program. For details, you must use the client of the alarm program to receive the message in real time. For details, see "alarm module and log ".

◆ Basic irrelevant

Keyboard: unknown scancode e0 5e

Undefined key values are received on the keyboard. If this occurs frequently, it may be caused by a keyboard problem. This is sometimes prompted for special keys or key combinations in linux. Check whether the server's keyboard is pressed. Otherwise, ignore it.

◆ Basic irrelevant

Uses obsolete (PF_INET, SOCK_PACKET)

The system kernel calls some functional modules, which may occur during the first call. This is generally related to the use of debugging tools.

◆ Network communication faults

Neighbour table overflow.

This prompt appears. Generally, some computers in the LAN are infected with viruses. In severe cases, communication may be affected. Computers with abnormal Intranet Communication must be processed.

◆ Network communication faults

Eth1: Transmit error, Tx status register 82. probably a duplex mismatch. see Documentation/networking/vortex.txt Flags; bus-master 1, dirty 9994190 (14) current 9994190 (14) Transmit list 00000000. f7171580. 0: @ f7171200 length 800001e6 status 000101e6 1: @ f7171240 length 8000008c status 0001008c ....

This prompt is available on the 3com Network. I feel that it will not be very serious if there is not a large volume. Currently, the solution for us to look at is to replace the NIC on the server. I really feel that 3com Nic has some problems...

◆ Serious server system faults

CPU0: Temperature above threshold CPU0: Running in modulated clock mode

The operating temperature of the server CPU is too high. hardware faults must be ruled out...