System and Network Administration

lecture in color

Providing services

without service, everything is easy.
set up system, start using it.
services make things more difficult
- special processes
- special initialization files

Where we are:

name of a service is triple (ip address, protocol, port number)
don't know how one associates a name with an actual service.

Providing services is a long story

structure of unix
processes and daemons
daemon scheduling

UNIX consists of four kinds of programs:

KERNEL - /boot/linix, the 'operating system'. responsible for:
- managing memory, disk, cpu time.
- scheduling tasks(processes)
SHELLS: /usr/bin/tcsh, /bin/sh (Bourne Shell)
- invoked by user actions.
- directly interact with users.
- control I/O for programs. a.out < file (glue file to input for a.out)
- translate CONTROL CHARACTERS into SIGNALS.
USER PROGRAMS - cat, gcc, etc.
- invoked by shells.
- directly interact with users.
- killed by control-C, etc passed from the SHELL.
DAEMONS - bootpd, named, etc.
- invoked at system startup
- end with 'd'.
- in */sbin or */etc.
- run as root(!)
- interact with OTHER SYSTEMS(and perhaps users thereupon).
- provide network services.
- a DAEMON is different from a DEMON. A DAEMON is a motivating (perhaps pantheistic) force, as in 'Maxwell's daemon'.
- controlled by use of SIGNALS.

Process concepts:

process ID: number between 0 and 65535 (wraps from 65535 to 0)
parent process: process that 'spawned' this one.
spawning:
- child inherits environment from parent
  - environment variables (PATH=, MANPATH=)
  - open files in parent stay open in child.

Forms of process spawning:

exec: replace current process with new one. In C:

      char *program, *arg1, *arg2, *arg3, ....; /* strings */ 
      execl(program, arg1, arg2, arg3, ....);

In csh,sh:

      exec /bin/cat

replaces the current shell with an instance of cat!

fork: make two COPIES of an existing process. in C:

      int childpid = fork(); 
      if (childpid) { /* in parent */ }
      else { /* in child */ }

fork and exec: make a copy that becomes another program. In C:

      int childpid = fork(); 
      if (childpid) { /* in parent */ }
      else { 
         execl(program, arg1, arg2, arg3, ...); 
      }

fork, exec, and join (fork-join): fork, exec, wait for completion. In C:

      system("/usr/bin/cat soandso");

or the equivalent:

      int childpid = fork(); 
      if (childpid) {
         int status = 0; 
         int gotpid = waitpid(childpid,&status,0); 
      } else { 
         /* parse system arguments into program name and arguments */ 
         execl("/usr/bin/cat", "soandso"); 
      }

Process spawning rules:

every process is a child of init (process 0).
init is its own parent.
every other process has a parent that spawned it.
manipulating a process automatically manipulates its immediate children
if a parent dies, the children are then on their own.
dirty trick (dissociation) to intentionally orphan a process, kill its parent with a fork! if (fork()) exit(0); Then the process has no living parent -> on its own.
if init dies, system reboots!

Viewing processes:

ps -ef (ps -aux) show all processes on a system.

 V user id
            v process id
                  v parent process id
                           v starting time 
                                    v device of invoking user
                                                 v cpu time used 
                                                         v command
 UID        PID  PPID    C STIME    TTY             TIME CMD
 root         0     0  0.1 12:55:10 ??           0:12.91 [kernel idle]
 root         1     0  0.0 12:55:10 ??           0:00.41 /sbin/init -a
 root         3     1  0.0 12:55:11 ??           0:00.29 /sbin/kloadsrv
 root        23     1  0.0 12:55:36 ??           0:01.55 /sbin/update
 root        96     1  0.0 12:55:44 ??           0:00.29 /usr/sbin/syslogd
 root        98     1  0.0 12:55:44 ??           0:00.04 /usr/sbin/binlogd
 root       238     1  0.0 12:55:51 ??           0:00.13 /sbin/named -b /etc/namedb/named.boot
 root       271     1  0.0 12:55:57 ??           0:00.16 /usr/sbin/portmap
 root       327     1  0.0 12:57:51 ??           0:00.13 -accepting connections (sendmail)
 root       371     1  0.0 12:57:53 ??           0:00.31 /usr/sbin/os_mibs
 root       373     1  0.0 12:57:53 ??           0:00.09 /usr/sbin/snmpd
 root       384     1  0.0 12:57:54 ??           0:00.11 /usr/sbin/inetd
 root       412     1  0.0 12:57:56 ??           0:00.15 /usr/sbin/joind
 root       419     1  0.0 12:57:57 ??           0:00.06 /usr/sbin/cron
 root       447     1  0.0 12:58:00 ??           0:00.03 /usr/lbin/lpd
 root       480     1  0.0 12:58:04 ??           0:01.42 /usr/dt/bin/dtlogin -daemon
 root       489   480  0.0 12:58:06 ??           1:39.79 /usr/bin/X11/X :0 -auth /var/dt/authdir/authfiles/A:0-aaapaa
 root       556   384  0.0 12:58:44 ??           0:00.49 rpc.ttdbserverd
 root      1283   480  0.0 13:42:18 ??           0:00.11 dtlogin <:0> -daemon
 couch     1307  1283  0.0 13:42:36 ??           0:02.45 /usr/dt/bin/dtsession
 couch     1346     1  0.0 13:42:48 ??           0:00.69 /usr/dt/bin/ttsession -s
 couch     1357  1307  0.0 13:42:59 ??           0:03.05 /usr/dt/bin/dthelpview -helpVolume Intromgr
 couch     1358  1307  0.0 13:42:59 ??           0:00.05 sh -c /usr/bin/X11/dxconsole
 couch     1359  1307  0.0 13:42:59 ??           0:04.53 dtfile -session dtaaazCa
 couch     1360  1307  0.0 13:42:52 ??           0:03.64 dtwm
 couch     1362  1364  0.0 13:43:07 ??           0:04.06 /usr/dt/bin/dtterm -ls
 root      1363  1358  0.0 13:42:59 ??           0:00.81 /usr/bin/X11/dxconsole
 couch     1364  1360  0.0 13:43:05 ??           0:00.33 /usr/dt/bin/dtexec -open 0 -ttprocid 2.qCJeO 01 1346 1342177279 1 0 30 192.168.1.1 3_101_1 /usr/dt/bin/dtterm -ls
 root      1380   384  0.4 14:15:09 ??           0:00.12 rlogind
 couch     1392  1393  0.0 14:14:06 ??           0:00.43 /usr/dt/bin/dtscreen -mode blank
 couch     1393  1307  0.0 14:14:06 ??           0:00.22 /usr/dt/bin/dtexec -open 0 -ttprocid 1.qCJeO 01 1346 1342177279 1 0 30 192.168.1.1 3_108_1 /usr/dt/bin/dtscreen -mode blank
 root       488     1  0.0 12:58:05 console      0:00.27 /usr/sbin/getty console console vt100
 couch     1369  1362  0.0 13:43:14 ttyp1        0:00.22 -csh (csh)
 couch     1389  1380  0.9 14:15:09 ttyp2        0:00.22 -csh (csh)
 root      1399  1389  2.8 14:15:15 ttyp2        0:00.15 ps -ef

Stupid process tricks:

It's often awkward to deal with processes directly.
use awk(1) or perl(1) to tear apart process LISTING.

list all process numbers owned by couch.

 ps -ef | awk '$1=="couch" { print $2 }'

kill all processes owned by couch:
```
 kill -9 `ps -ef | awk '$1=="couch" { print $2 }'`
```
(substitutes the output of the former into the command)
kill all commands called emacs (including commands stupid enough to name themselves emacs without being emacs:
```
 kill -9 `ps -ef | awk '$8 ~ /emacs/ { print $2 }'`
```

Awk:

table processing language
complete programs occupy one line.
each program has a guard clause and an action clause.
variables $1,$2,$3 apply to fields in input.
guard clauses:
- $8 ~ /emacs/: matches any line in which field 8 CONTAINS the pattern emacs.
- $1 == "couch" : equality. >=, >, <=, <, != similarly.
- $8 !~ /emacs/ : matches anything that doesn't contain emacs.
- A && B, A || B, etc. all work as in C.
program statements:
- print something; prints the thing.
- for, while work as in C.
- variables are created when referenced.
- Arrays are full associative: a["foo"]="bar"; is legal.

EX: simple program to figure out how many processes people have:

 BEGIN {       # BEGIN is done before any input read. 
     count=0; 
 }
 $2 != "PID" { # skip first line of ps. 
     # for each new UID, put into people. 
     # then increment procs[uid]. 
     if (!procs[$1]) { people[count++]=$1; }
     procs[$1]++;      # count for each PERSON of processes. 
 }
 END {                 # END is done after eof on input. 
     for (i=0; i<count; i++) { 
       print people[i],procs[people[i]]; 
     }
 }
 
 ps -ef | awk -f count.awk

runs this if it's in count.awk
result:
```
 root 24
 bmatijas 12
 couch 3
```

Controlling processes:

when running under a shell, processes respond to CONTROL CHARACTERS. These are translated by the shell into SIGNALS sent to the process.
when running without a controlling terminal, processes communicate via SIGNALS. A SIGNAL is a message containing an integer. Each integer has a specific meaning.

In /usr/include/sys/signal.h:

 #define SIGHUP     1    /* hangup, generated when terminal disconnects */
 #define SIGINT     2    /* interrupt, generated from terminal special char */
 #define SIGQUIT    3    /* (*) quit, generated from terminal special char */
 #define SIGILL     4    /* (*) illegal instruction (not reset when caught)*/
 #define SIGTRAP    5    /* (*) trace trap (not reset when caught) */
 #define SIGABRT    6    /* (*) abort process */
 #define SIGEMT     7    /* (*) EMT instruction */
 #define SIGFPE     8    /* (*) floating point exception */
 #define SIGKILL    9    /* kill (cannot be caught or ignored) */
 #define SIGBUS    10    /* (*) bus error (specification exception) */
 #define SIGSEGV   11    /* (*) segmentation violation */
 #define SIGSYS    12    /* (*) bad argument to system call */
 #define SIGPIPE   13    /* write on a pipe with no one to read it */
 #define SIGALRM   14    /* alarm clock timeout */
 #define SIGTERM   15    /* software termination signal */
 #define SIGURG    16    /* (+) urgent contition on I/O channel */
 #define SIGSTOP   17    /* (@) stop (cannot be caught or ignored) */
 #define SIGTSTP   18    /* (@) interactive stop */
 #define SIGCONT   19    /* (!) continue if stopped */
 #define SIGCHLD   20    /* (+) sent to parent on child stop or exit */
 #define SIGTTIN   21    /* (@) background read attempted from control terminal*/
 #define SIGTTOU   22    /* (@) background write attempted to control terminal */
 #define SIGPOLL   23    /* I/O possible, or completed */
 #define SIGXCPU   24    /* cpu time limit exceeded (see setrlimit()) */
 #define SIGXFSZ   25    /* file size limit exceeded (see setrlimit()) */
 #define SIGVTALRM 26    /* virtual time alarm (see setitimer) */
 #define SIGPROF   27    /* profiling time alarm (see setitimer) */
 #define SIGWINCH  28    /* (+) window size changed */
 #define SIGINFO   29    /* (+) information request */
 #define SIGUSR1   30    /* user defined signal 1 */
 #define SIGUSR2   31    /* user defined signal 2 */
 #define SIGRESV   32    /* reserved by Digital for future use */

'kill' doesn't just kill programs, but sends 'signals' to programs:

   kill -HUP {pid}  # kill -1 {pid} 
   kill -KILL {pid} # kill -9 {pid}

Special signals:

INT (2) die with full honors (control-C).
QUIT (3) die with core dump (control-\).
KILL (9) die without honor.
TERM (15) kindest kind of death. required for processes that must clean up after themselves (automountd)
HUP (1) re-read databases (VERY IMPORTANT).

Signal handling:

some processes BLOCK some signals.
must remember that SIGINT and SIGQUIT are blockable, but SIGKILL isn't.

Configuring a service:

edit its databases.
start its daemon(s).
arrange for them to start at system startup.

When daemons are started: run-levels:

BSD system: two run levels (single user and multiuser)
- single-user: no network services. (control-D -> multiuser)
- multiuser: network services. (shutdown -> single user)
System V: multiple 'run levels'. 0: dead. 1, S: single user. 2: user services (interactive login). 3: network services (nfs, etc) 6: reboot.
Linux: 5 run levels, ill-defined 0: dead 1: single-user 5: multi-user with network services enabled.
to change run levels: init 3 ( go to multiuser) init 0 # /etc/halt init 6 # /etc/reboot

Two ways to start services:

BSD:
- /etc/rc.boot, rc.single: bring up single user mode from nothing.
- /etc/rc.local: command script to run at beginning of multiuser mode.
- sh scripts: bizarre syntax
System V: (OSF4)
/etc/init.d, /sbin/init.d: lots of little individual scripts to run.
- each script takes two possible parameters, 'stop' and 'start'
- 'start' : start a daemon
- 'stop' : stop a daemon
- /sbin/init.d/named: script that starts and stops name service.
  - /sbin/init.d/named start : starts it.
  - /sbin/init.d/named stop : stops it.

/sbin/rc*.d: ordered list of scripts to run

symlinks to things in /sbin/init.d

    /sbin/rc3.d/S15named -> /sbin/init.d/named MEANS
            ^ run level 
                ^^^ Action and order (lower is earlier) 
                   ^^^^^ daemon to start or stop. S=start, K=stop.

Call this with parameter start, when system coming up to run level 3.

    /sbin/rc2.d/K35nfs -> /sbin/init.d/nfs MEANS
            ^ run level 
                ^^^ Action and order (lower is earlier)

Call this with parameter stop, when system coming down past run level 2.

Linux: /etc/rc.d/init.d, /etc/rc.d/rc?.d.

Reconfiguring a service:

edit databases.
tell daemons to re-read databases by
1. kill and restart (lpd, lpsched, etc. )
  - lpd : ps -ef | grep lpd; kill {pid} ; /usr/lib/lpd.
  - lpsched: /usr/bin/lpshut ; /usr/lib/lpsched
2. send HUP (bootp, inetd, named)
  - if doesn't work the first time, might have to kill and restart.
3. kill with auto-restart (inetd daemons, in.*)
  - bootpd will restart itself if you kill it!
4. send SIGTERM and restart (automount).
  - things get REALLY AWFUL if you don't.
5. reboot (automount in practice).

Must remember what to do for each specific daemon

Classes of daemons:

organic: run by themselves, run all the time, started at boot time.
embedded: run by the superdaemon _inetd_.
- wait: run only one instance when service requested.
  - typically, this daemon will lock into memory,
  - to restart, simply kill the instance.
  - EX: bootp.
  - fork, exec, join of inetd.
  - Daemon itself has control over further forking (forking consumes system resources)
- nowait: run as many instances as demands require.
  - typically, daemon responds and exits.
  - inetd then schedules another.
  - no restart needed.
  - fork and exec.
  - inetd fails to control forking.

Inetd configuration

/etc/inetd.conf: configuration file for inetd.
inetd listens on ALL PORTS LISTED IN inetd.conf.
selectively runs daemons as needed.

EX: BOOTP has a line:

 #name  socktyp proto   wait    user    command                 arguments
 bootps dgram   udp     wait    root    /usr/sbin/bootpd        bootpd
 ^ SERVICE NAME (from /etc/services) 
        ^ SOCKET TYPE: dgram, stream, raw, rdm, seqpacket. 
                ^ PROTOCOL: udp, tcp, or any other protocol in /etc/protocols. 
                        ^ wait or nowait. 
                              ^ user to run this as
                                      ^ COMMAND to execute

Sample inetd file

 # Syntax for socket-based Internet services:
 #  <service_name> <socket_type> <proto> <flags> <user> <server_pathname> <args>
 #
 # Syntax for TLI-based Internet services:
 #  <service_name> tli <proto> <flags> <user> <server_pathname> <args>
 #
 # Ftp and telnet are standard Internet services.
 #
 ftp   stream  tcp     nowait  root    /usr/sbin/tcpd  in.ftpd
 telnet        stream  tcp     nowait  root    /usr/sbin/tcpd  in.telnetd
 #
 # Tnamed serves the obsolete IEN-116 name server protocol.
 #
 name  dgram   udp     wait    root    /usr/sbin/tcpd  in.tnamed
 #
 # Shell, login, exec, comsat and talk are BSD protocols.
 #
 shell stream  tcp     nowait  root    /usr/sbin/tcpd  in.rshd
 login stream  tcp     nowait  root    /usr/sbin/tcpd  in.rlogind
 exec  stream  tcp     nowait  root    /usr/sbin/tcpd  in.rexecd
 comsat        dgram   udp     wait    root    /usr/sbin/tcpd  in.comsat
 talk  dgram   udp     wait    root    /usr/sbin/tcpd  in.talkd
 #
 # Must run as root (to read /etc/shadow); "-n" turns off logging in utmp/wtmp.
 #
 uucp  stream  tcp     nowait  root    /usr/sbin/tcpd  in.uucpd
 #
 # Tftp service is provided primarily for booting.  Most sites run this
 # only on machines acting as "boot servers." 
 #
 #tftp dgram   udp     wait    root    /usr/sbin/in.tftpd      in.tftpd -s /tftpboot
 #
 # Finger, systat and netstat give out user information which may be
 # valuable to potential "system crackers."  Many sites choose to disable 
 # some or all of these services to improve security.
 #
 finger        stream  tcp     nowait  nobody  /usr/sbin/tcpd  in.fingerd
 #systat       stream  tcp     nowait  root    /usr/bin/ps             ps -ef
 #netstat      stream  tcp     nowait  root    /usr/bin/netstat        netstat -f inet
 #
 # Time service is used for clock synchronization.
 #
 time  stream  tcp     nowait  root    internal
 time  dgram   udp     wait    root    internal
 # 
 # Echo, discard, daytime, and chargen are used primarily for testing.
 #
 echo  stream  tcp     nowait  root    internal
 echo  dgram   udp     wait    root    internal
 discard       stream  tcp     nowait  root    internal
 discard       dgram   udp     wait    root    internal
 daytime       stream  tcp     nowait  root    internal
 daytime       dgram   udp     wait    root    internal
 chargen       stream  tcp     nowait  root    internal
 chargen       dgram   udp     wait    root    internal
 ...

Recall /etc/services: what are network services?

 tcpmux                1/tcp
 echo          7/tcp
 echo          7/udp
 discard               9/tcp           sink null
 discard               9/udp           sink null
 systat                11/tcp          users
 daytime               13/tcp
 daytime               13/udp
 netstat               15/tcp
 chargen               19/tcp          ttytst source
 chargen               19/udp          ttytst source
 ftp-data      20/tcp
 ftp           21/tcp
 telnet                23/tcp
 smtp          25/tcp          mail
 time          37/tcp          timserver
 time          37/udp          timserver
 name          42/udp          nameserver
 whois         43/tcp          nicname         # usually to sri-nic
 domain                53/udp
 domain                53/tcp
 bootps                67/udp                          # BOOTP/DHCP server
 bootpc                68/udp                          # BOOTP/DHCP client
 hostnames     101/tcp         hostname        # usually to sri-nic
 pop2          109/tcp         pop-2           # Post Office Protocol - V2
 pop3          110/tcp                         # Post Office Protocol - Version 3
 sunrpc                111/udp         rpcbind
 sunrpc                111/tcp         rpcbind
 imap          143/tcp         imap2           # Internet Mail Access Protocol v2
 ldap          389/tcp                         # Lightweight Directory Access Protocol 
 ldap          389/udp                         # Lightweight Directory Access Protocol
 ldaps         636/tcp                         # LDAP protocol over TLS/SSL (was sldap)
 ldaps         636/udp                         # LDAP protocol over TLS/SSL (was sldap)
 #
 # Host specific functions
 #
 tftp          69/udp
 rje           77/tcp
 finger                79/tcp
 link          87/tcp          ttylink
 supdup                95/tcp
 iso-tsap      102/tcp
 x400          103/tcp                         # ISO Mail
 x400-snd      104/tcp
 csnet-ns      105/tcp
 pop-2         109/tcp                         # Post Office
 uucp-path     117/tcp
 nntp            119/tcp         usenet                # Network News Transfer
 ntp           123/tcp                         # Network Time Protocol
 ntp           123/udp                         # Network Time Protocol
 NeWS          144/tcp         news            # Window System
 cvc_hostd     442/tcp                         # Network Console
 #
 # UNIX specific services
 #
 # these are NOT officially assigned
 #
 exec          512/tcp
 login         513/tcp
 shell         514/tcp         cmd             # no passwords used
 printer               515/tcp         spooler         # line printer spooler
 courier               530/tcp         rpc             # experimental
 uucp          540/tcp         uucpd           # uucp daemon
 biff          512/udp         comsat
 who           513/udp         whod
 syslog                514/udp
 talk          517/udp
 route         520/udp         router routed
 klogin                543/tcp                         # Kerberos authenticated rlogin
 new-rwho      550/udp         new-who         # experimental
 rmonitor      560/udp         rmonitord       # experimental
 monitor               561/udp                         # experimental
 pcserver      600/tcp                         # ECD Integrated PC board srvr
 kerberos-adm  749/tcp                         # Kerberos V5 Administration
 kerberos-adm  749/udp                         # Kerberos V5 Administration
 kerberos      750/udp         kdc             # Kerberos key server
 kerberos      750/tcp         kdc             # Kerberos key server
 krb5_prop     754/tcp                         # Kerberos V5 KDC propogation
 ufsd          1008/tcp        ufsd            # UFS-aware server
 ufsd          1008/udp        ufsd
 cvc           1495/tcp                        # Network Console
 ingreslock      1524/tcp
 www-ldap-gw   1760/tcp                        # HTTP to LDAP gateway
 www-ldap-gw   1760/udp                        # HTTP to LDAP gateway
 listen          2766/tcp                        # System V listener port
 nfsd          2049/udp        nfs             # NFS server daemon (clts)
 nfsd          2049/tcp        nfs             # NFS server daemon (cots)
 eklogin               2105/tcp                        # Kerberos encrypted rlogin
 lockd         4045/udp                        # NFS lock daemon/manager
 lockd         4045/tcp
 dtspc         6112/tcp                        # CDE subprocess control
 fs            7100/tcp                        # Font server
 ...

/etc/protocols: what are names of protocols?

 ip            0       IP              # internet protocol, pseudo protocol number
 icmp          1       ICMP            # internet control message protocol
 ggp           3       GGP             # gateway-gateway protocol
 tcp           6       TCP             # transmission control protocol
 egp           8       EGP             # exterior gateway protocol
 pup           12      PUP             # PARC universal packet protocol
 udp           17      UDP             # user datagram protocol
 hmp           20      HMP             # host monitoring protocol
 xns-idp               22      XNS-IDP         # Xerox NS IDP
 rdp           27      RDP             # "reliable datagram" protocol
 
 #
 # Internet (IPv6) extension headers
 #
 ipv6          41      IPv6            # IPv6 in IP encapsulation
 ipv6-route    43      IPv6-Route      # Routing header for IPv6
 ipv6-frag     44      IPv6-Frag       # Fragment header for IPv6
 esp           50      ESP             # Encap Security Payload for IPv6
 ah            51      AH              # Authentication Header for IPv6
 ipv6-icmp     58      IPv6-ICMP       # IPv6 internet control message protocol
 ipv6-nonxt    59      IPv6-NoNxt      # No next header extension header for IPv6
 ipv6-opts     60      IPv6-Opts       # Destination Options for IPv6

Inetd function:

pretend to be all services listed in /etc/inetd.conf.
if someone contacts you, fork and exec the CORRECT daemon and let it respond.
this works because:
- forked programs have the same open files as the original
- so this one inherits a connection to the requestor
- so it can pretend it was running all along.

How to read inetd.conf

actually need to understand several files

inetd.conf: what inetd services are.

 telnet        stream  tcp     nowait  root    /usr/sbin/tcpd  in.telnetd
                       ^^^proto^^^^^^execution control
 ^^^^^^ service^^^^^^kind              ^^^^ user               ^^^^^^^^^^ progname
                                               ^^^^^^^^^^^^^^file

services: explains field 1 of inetd.conf

 telnet                23/tcp
 ^^^^^^ service name   ^^number
                          ^^^ protocol (/etc/protocols)

protocols: explains field 3 of inetd.conf

 tcp           6       TCP             # transmission control protocol
 ^^^ name      ^ number

The awful truth: names are fake

none of these names are known to the Internet
They're provided for HUMANS
Actual packets contain the NUMBERS
- TCP -> 6
- telnet -> 23
to read raw packets, must translate between names and numbers!

Two kinds of inetd daemons

wait: daemon terminates at completion of each request.
- called this because inetd must call the wait(2) system call.
- you don't have to kill these
nowait: daemon stays active until requests cease.
- you may have to kill these if they don't die.

Time/space tradeoff

daemons that run all the time
- take memory all the time
- respond to requests quickly.
- ex httpd (web service)
daemons that are started by inetd
- don't run all the time
- take no memory when idle
- ex: telnetd

lecture in color

/comp/150NET/notes/service-old.php
downloaded on Mar-16-2010 04:47:40 PM,
was last modified on Feb-17-2004 10:49:51 PM.
All lecture note content is copyright 2004 by
Alva L. Couch, Computer Science, Tufts University