×

Loading...
Ad by
  • 推荐 OXIO 加拿大高速网络,最低月费仅$40. 使用推荐码 RCR37MB 可获得一个月的免费服务
Ad by
  • 推荐 OXIO 加拿大高速网络,最低月费仅$40. 使用推荐码 RCR37MB 可获得一个月的免费服务

I also met some system crash on Linux, most time It was some hardware problem. some time it was some misconfiged software. of couse there were aslo some case it dead for unknown reason .

But without enough evidence, I wouldn't say It's Linux's unstablity caused machine halt. Because most machines just work fine. I think the present Linux should be stable enough for commercial purpose. Currently many ISPs have already chose Linux as their main unix type webhosting server (e.g Godaddy, I bet they have more than several hundreds Linux server. )

Compared with other unix type OS, Linux has much more bigger user and developer group. The good thing about it is Linux support much more hardware resource than other unix, because there are so many people writting driver on linux. That means you have more options to choose. The bad thing is not all the drivers are well develop/tested. So if unfornately you made a bad choice, it's reasonable the system is not that stable. On ther other hand, some other os ( like Sun solaris?) they can focus on a much narrowed hardware list.
Report

Replies, comments and Discussions:

  • 工作学习 / 专业技术讨论 / 请教一下,有台SUSE SERVER,PING所有的网段都通,就是所有的USER都LOGIN不了, 啥毛病啊?
    • telnet and ping are using different protocol type. Check out if telnet is enable on this server. Check the user manual to find out how to enable it.
      • 都设了,一直用得好好的. 突然就不行了. REBOOT以后就好了. 要写事故报告,不知道什么原因~~~~
        • 难说,最坏的情况是一个黑客潜入,在你的机器里埋下了一个程序。专门截获你们的login信息,但是该程序必须重新启动才能生效。before, U -> M, 现在,U -> H -> M. 详细间内。
          开个玩笑 -:)。不知道了。得看Admin Users Guide
        • 你先搞清楚是用SSH还是Telnet再说
          ssh是由sshd daemon 提供连接的, telnet 是由telnetd 来提供连接的.
          基本上无需重启机器, 只要重启这两个daemon 之一就好了.
          原因嘛, 就是Linux 的不稳定性 :)))
          • There are more than 20 Linux machines in my company, RH8, RH9, FC, Debian, Suse... . I never saw one sshd daemon down because of "the unstable Linux OS".
            • 是没有显示Login的提示还是输入密码不接受?
            • make a cron job monitor all the processes for several days. ps -ef >> pro.log And make sure all processes are the ones suppose to run.
              • 是谁都不能ACCESS, 连ROOT都不行, 啥也看不到,只好POWER RESET.
                • Monitor the telnetd or sshd by the cron job. Check your telnetd or sshd with other machine. Check the file size and date to see if there was any thing abnormal. If the size is wrong, then it is a hacker program.
              • also insert this line in the cron job. top >> res.log It will monitor your resouce usage. If it happens next time, you will know what happened. You can use this method for all your machine if you are the admin
              • forgot mention. Make sure you cron job output will not jam your disk. You can use the date as the part of the output file name and delete the old date's file when writing to new one
            • You guys are lucky . My company has several hundreds Linux servers.
              Sometimes some are halt without any reason,even we can not logon from console(need a power reset) , some's sshd/telnetd daemon do not response need to restart sshd or telnetd /xinetd.
              We only consider the development boxes, non- mission critical servers to LINUX. All missison critical servers are commercial UNIX. Nobody even dare to use LINUX in these area.
              I don't hate LINUX, but I wish we had the same "stable" LINUX as yours, I wish LINUX will be more stable in one or two years.
              • I also met some system crash on Linux, most time It was some hardware problem. some time it was some misconfiged software. of couse there were aslo some case it dead for unknown reason .
                But without enough evidence, I wouldn't say It's Linux's unstablity caused machine halt. Because most machines just work fine. I think the present Linux should be stable enough for commercial purpose. Currently many ISPs have already chose Linux as their main unix type webhosting server (e.g Godaddy, I bet they have more than several hundreds Linux server. )

                Compared with other unix type OS, Linux has much more bigger user and developer group. The good thing about it is Linux support much more hardware resource than other unix, because there are so many people writting driver on linux. That means you have more options to choose. The bad thing is not all the drivers are well develop/tested. So if unfornately you made a bad choice, it's reasonable the system is not that stable. On ther other hand, some other os ( like Sun solaris?) they can focus on a much narrowed hardware list.
    • The problem is the login process. What athetication method do you use now? I have run into the problem when I configure my openssh
      • 那台SERVER一直好好的,就突然不能LOGIN了, 我把它REBOOT了以后就好了,可是为什么会发生这种情况呢? 去哪里查LOG?
        • How long has it been run? It's possible that ur SWAP space was full. which would lead to no 'New session' can be forked.
          • also the /opt, if it is full, system would crash without warning. /opt can be checked now. But should not be this reason. She said it was working now
        • what login process you are using? rlogin, telnet or ssh ?
    • next time, setup serial line, use a dumb terminal or hyperTerminal to access root. if it is OK:net service has problem. NO:login it self has problem.