Facultat d'informàtica de Barcelona Univ. Politècnica de Catalunya. Administració de Sistemes Operatius. System monitoring

Transcription

1 Facultat d'informàtica de Barcelona Univ. Politècnica de Catalunya Administració de Sistemes Operatius System monitoring

2 Topics 1. Introduction to OS administration 2. Installation of the OS 3. Users management 4. Applications management 5. System monitoring 6. Maintenance of the file system 7. Local services 8. Network services 9. Protection and security

3 Knowledge Objectives Commands and tools for system monitoring Meaning of each inter-process signals Abilities Obtain information about the system state CPU activity Memory activity Disc activity Change the state of processes Priority settings Stop and resume processes

4 Monitoring Why should we monitor the system? Have a control on the use of resources pro-active, well in advance of problems Control the state of services Protection and security Actions Automatic Manual

5 Monitoring What should we monitor? CPU Memory I/O Network Users Services Logs

6 Monitoring When should we start monitoring a resource? Who should be notified when there is a problem? What criteria should be used to notify a warning? And to notify a critical problem?

7 CPU activity Monitor Idle processors Monopolized processors Tools By a single process By a single user uptime, top, ps

8 Memory activity Monitor Memory shortage Monopolized memory By a single process By a single user Swap area Tools free, vmstat, top

9 Disc activity Monitor File system Anomalous I/O activity Swap space activity Excess of paging Free memory available Tools vmstat, df, iostat

10 Network activity Monitor Communication bandwidth Local and remote services Input/output connections Tools ifconfig, netstat, tcpdump, nmap, logs del sistema

11 Users Monitor Active sessions Locally Remotely Connected users Tools What are they doing? w, last, finger, fuser, lsof

12 Other monitoring tasks Servers & services activity Web server load queues Incoming Outgoing Printer queues Log files System errors Anomalous activity (security)

13 Tasks related to process management Identify the process Which user is the owner of the process? Which task is it performing? How important is it? Is this an attack?... or an error? Manage the process appropriately Change its priority Stop and resume the process Kill the process

14 Managing priorities When executing the process nice +10 command... While the process is running renice +10 <pid> Only root can increase priorities Negative values indicate higher priorities

15 High priority shell An advice... When the system load is high, a high priority shell can help to investigate what is happening Children processes inherit parent priority

16 Send signals to a process kill <signal> <pid> -KILL: process ends with no option to continue -TERM: asks the process to finish (by default, it kills) -INT: interrupt the process (by default, it kills) -STOP: stop a process Cannot enter the ready queue while stopped -CONT: resume a stopped process killall <signal> <command name> Sends the signal to all processes in the system executing the indicated command

17 User activity w [user] User monitoring Lists connected users and the command they are executing With a username, it lists only the connections of him/her last [user] Lists the last connections established to the machine finger [user] finished or not Lists all connections, or those of the given user

18 User monitoring File activity fuser <filename> Identifies processes that are using a specified file lsof [filename dirname] Lists processes that have the file opened, or that are inside the directory

19 Disc monitoring Used space du [filename dirname] (disk usage) Indicates the space used by a file or directory (and its descendents) Free space df [filename dirname] (disk free) Available disk space in the partition where the file resides I/O activity vmstat iostat

20 top 4:50pm up 11 days, 8:23, 7 users, load average: 0.01, 0.06, processes: 126 sleeping, 1 running, 1 zombie, 0 stopped CPU0 states: 0.1% user, 0.0% system, 0.0% nice, 99.4% idle CPU1 states: 1.0% user, 0.0% system, 1.0% nice, 98.4% idle CPU2 states: 0.1% user, 1.4% system, 0.0% nice, 97.4% idle CPU3 states: 0.0% user, 0.0% system, 0.0% nice, 100.0% idle Mem: K av, K used, 36272K free, 0K shrd, 88516K buff Swap: K av, 52560K used, K free K cached PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND 10 root SWN :40 kscand/highmem pareta M 120M S N :43 mozilla-bin admac-e M 3676 S N :10 mysqld pareta M 120M S :00 mozilla-bin aduran R :00 top 1 root S :11 init 2 root SW :03 keventd 3 root SWN :00 ksoftirqd_cpu0 4 root SWN :00 ksoftirqd_cpu1 5 root SWN :00 ksoftirqd_cpu2 6 root SWN :00 ksoftirqd_cpu3 7 root SW :40 kswapd 8 root SW :11 kscand/dma 9 root SWN :44 kscand/normal 11 root SW :04 bdflush 12 root SW :17 kupdated 13 root SW< :00 mdrecoveryd 17 root SW :30 kjournald 96 root SW :00 khubd

21 vmstat # vmstat -n 30 procs memory swap io---- -system cpu---- r b swpd free buff cache si so bi bo in cs us sy id wa

22 Activity Which problem do you think it happens in this server? Which actions would you take? top - 17:10:26 up 11 days, 8:33, 2 users, load average: 2.65, 1.22, 0.48 Tasks: 70 total, 4 running, 66 sleeping, 0 stopped, 0 zombie Cpu0 : 48.2%us, 0.4%sy, 0.0%ni, 51.4%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Mem: k total, k used, 6268k free, 49984k buffers Swap: k total, 44k used, k free, 50644k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND aduran R :15.23 updatesw aduran R :38.99 merge aduran R :29.82 merge aduran R :03.77 top 1 root S :02.89 init 2 root RT S :00.00 migration/0 3 root S :00.06 ksoftirqd/0 4 root S :00.02 events/0 5 root S :00.01 khelper 6 root S :00.00 kthread 9 root S :00.09 kblockd/0 10 root S :00.00 kacpid 66 root S :00.00 kseriod 100 root S :00.01 pdflush 101 root S :03.75 pdflush 102 root S :04.67 kswapd0 103 root S :00.00 aio/0

23 Activity Which problem do you think it happens in this server? Propose a solution top - 00:39:54 up 41 days, 14:53, 3 users, load average: 2.49, 0.98, 0.36 Tasks: 66 total, 1 running, 65 sleeping, 0 stopped, 0 zombie Cpu(s): 0.7%us, 10.3%sy, 0.0%ni, 50.3%id, 37.7%wa, 1.0%hi, 0.0%si, 0.0%st Mem: k total, k used, 3556k free, 760k buffers Swap: k total, k used, k free, 1876k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 8818 aduran m 86m 68 S :02.00 compact 96 root S :29.44 kswapd0 777 xavim m 81m 68 S :07.74 netscape 877 root R :01.31 top 1 root S :05.77 init 2 root RT S :00.00 migration/0 4 root S :00.02 events/0 5 root S :00.01 khelper 6 root S :00.00 kthread 9 root S :00.09 kblockd/0 10 root S :00.00 kacpid 66 root S :00.00 kseriod 100 root S :00.01 pdflush 101 root S :03.75 pdflush 102 root S :04.67 kswapd0 103 root S :00.00 aio/0

24 Integrated systems Network monitoring Centralize the information from several servers Resources Services Uptime Connectivity Logs Make it easy the detection of problems NagiOS, Splunk

25 NagiOS

26 NagiOS

27 Personal work Backup tools dump tar gzip, bzip2, zip, rar, partimage, Norton Ghost