Ƙirƙirar Tsarukan Fayil ɗin Fayil na Linux da Daidaita Sabar NFSv4 - Kashi na 2

An horar da Injiniya Takaddar Gidauniyar Linux (LFCE) don saitawa, daidaitawa, sarrafawa, da magance ayyukan cibiyar sadarwa a cikin tsarin Linux, kuma ana ba da amsa ga ƙira da aiwatar da tsarin gine-gine da warware matsalolin yau da kullun.

Gabatar da Shirin Takaddun Shaida na Gidauniyar Linux (LFCE).

A cikin Sashe na 1 na wannan silsilar mun bayyana yadda ake shigar da uwar garken NFS (Network File System), da kuma saita sabis ɗin don farawa ta atomatik akan boot. Idan baku riga kuka yi haka ba, da fatan za a koma ga wannan labarin kuma ku bi matakan da aka zayyana kafin a ci gaba.

  1. Shigar da Sabis na Yanar Gizo da Haɓaka Farawa ta atomatik a Boot - Part 1

Yanzu zan nuna muku yadda ake daidaita sabar ku ta NFSv4 da kyau (ba tare da ingantaccen tsaro ba) ta yadda zaku iya saita hannun jarin hanyar sadarwa don amfani da abokan cinikin Linux kamar an shigar da waɗancan tsarin fayiloli a cikin gida. Lura cewa zaku iya amfani da LDAP ko NIS don dalilai na tantancewa, amma duka zaɓuɓɓukan biyun sun fita daga iyakokin takaddun shaida na LFCE.

Ana saita uwar garken NFSv4

Da zarar uwar garken NFS ya tashi yana aiki, za mu mai da hankali kan:

  1. bayyana da daidaita kundayen adireshi na gida waɗanda muke son rabawa akan hanyar sadarwar, da
  2. Haɓaka waɗancan hannun jarin hanyar sadarwa a cikin abokan ciniki ta atomatik, ko dai ta hanyar fayil ɗin /etc/fstab ko kuma kayan aiki na tushen kernel (autofs).

Za mu yi bayanin daga baya lokacin da za a zaɓi hanya ɗaya ko ɗaya.

Kafin mu kasance, muna buƙatar tabbatar da cewa idmapd daemon yana gudana kuma yana daidaita shi. Wannan sabis ɗin yana aiwatar da taswirar NFSv4 sunaye ([email kare ) zuwa ID na mai amfani da rukuni, kuma ana buƙatar aiwatar da sabar NFSv4.

Shirya /etc/default/nfs-common don kunna idmapd.

NEED_IDMAPD=YES

Kuma gyara /etc/idmapd.conf tare da sunan yankin ku (tsofaffin shine FQDN na mai watsa shiri).

Domain = yourdomain.com

Sannan fara idmapd.

# service nfs-common start 	[sysvinit / upstart based systems]
# systemctl start nfs-common 	[systemd based systems]

Fayil na /etc/exports yana ƙunshe da manyan umarnin daidaitawa don uwar garken NFS ɗinmu, yana bayyana tsarin fayil ɗin da za a fitar dashi zuwa runduna mai nisa kuma yana ƙayyadad da zaɓuɓɓukan da ake da su. A cikin wannan fayil ɗin, ana nuna kowane rabon hanyar sadarwa ta amfani da keɓantaccen layi, wanda ke da tsari mai zuwa ta tsohuwa:

/filesystem/to/export client1([options]) clientN([options])

Inda /filesystem/to/fitarwa shine cikakkiyar hanyar zuwa tsarin fayil ɗin da aka fitar, yayin da client1 (har zuwa clientN) yana wakiltar takamaiman abokin ciniki (sunan mai masauki ko adireshin IP) ko hanyar sadarwa (ana ba da izinin kati) wanda rabon da ake fitarwa zuwa waje. A ƙarshe, zaɓuɓɓuka shine jerin ƙimar waƙafi (zaɓuɓɓuka) waɗanda ake la'akari yayin fitar da rabon, bi da bi. Da fatan za a lura cewa babu sarari tsakanin kowane sunan mai masauki da baƙaƙen da yake gaba da shi.

Anan akwai jerin zaɓuɓɓukan mafi yawan lokuta da bayanin su:

  1. ro (gajeren don karantawa kawai): Abokan ciniki masu nisa na iya hawa tsarin fayil ɗin da aka fitar tare da izinin karantawa kawai.
  2. rw (gajeren karanta-rubutu): Yana ba da damar runduna ta nesa su yi canje-canje a cikin tsarin fayil ɗin da aka fitar.
  3. wdelay (gajeren jinkirin rubutu): Sabar NFS tana jinkirta yin canje-canje zuwa faifai idan tana zargin wani buƙatun rubutu mai alaƙa yana nan kusa. Koyaya, idan uwar garken NFS ta karɓi ƙananan buƙatun da ba su da alaƙa da yawa, wannan zaɓin zai rage aiki, don haka za a iya amfani da zaɓin no_wdelay don kashe shi.
  4. sync: Sabar NFS tana amsa buƙatun ne kawai bayan an ƙaddamar da canje-canje zuwa ma'ajiyar dindindin (watau hard disk). Akasin sa, zaɓin async, na iya ƙara aiki amma a farashin asarar bayanai ko ɓarna bayan sabar mara tsabta ta sake farawa.
  5. root_squash: Yana hana masu amfani da tushen nesa samun gata na masu amfani a cikin uwar garken kuma ya sanya musu ID na mai amfani ga kowa. Idan kuna son \squash duk masu amfani (ba tushen kawai ba), kuna iya amfani da zaɓin all_squash.
  6. annuid/anongid: A bayyane yake saita UID da GID na asusun da ba a san shi ba (ba kowa).
  7. subtree_check: Idan ƙaramin kundin tsarin fayil ne kawai aka fitar da shi, wannan zaɓin yana tabbatar da cewa fayil ɗin da aka nema yana cikin wannan ƙaramin directory ɗin da aka fitar. A gefe guda, idan an fitar da dukkan tsarin fayil zuwa waje, kashe wannan zaɓi tare da no_subtree_check zai hanzarta canja wurin. Zaɓin da aka saba a zamanin yau shine no_subtree_check saboda duban bishiyar yana haifar da ƙarin matsaloli fiye da yadda ya dace, bisa ga fitar da man 5.
  8. fsid=0 | tushen (sifili ko tushen): Yana ƙayyade cewa ƙayyadadden tsarin fayil shine tushen kundayen adireshi da yawa da aka fitar (kawai a cikin NFSv4).

A cikin wannan labarin za mu yi amfani da kundayen adireshi /NFS-SHARE da /NFS-SHARE/mydir akan 192.168.0.10 (Sabar NFS) a matsayin mu gwajin tsarin fayil.

A koyaushe muna iya lissafin samammun hannun jari na cibiyar sadarwa a cikin sabar NFS ta amfani da umarni mai zuwa:

# showmount -e [IP or hostname]

A cikin abubuwan da ke sama, za mu iya ganin cewa an fitar da /NFS-SHARE da /NFS-SHARE/mydir hannun jari akan 192.168.0.10 zuwa abokin ciniki mai adireshin IP 192.168.0.17.

Tsarin mu na farko (koma zuwa /da sauransu/fitarwa directory akan sabar NFS ku) don kundin da aka fitar dashi kamar haka:

/NFS-SHARE  	192.168.0.17(fsid=0,no_subtree_check,rw,root_squash,sync,anonuid=1000,anongid=1000)
/NFS-SHARE/mydir    	192.168.0.17(ro,sync,no_subtree_check)

Bayan gyara fayil ɗin sanyi, dole ne mu sake kunna sabis na NFS:

# service nfs-kernel-server restart 		[sysvinit / upstart based system]
# systemctl restart nfs-server			[systemd based systems]

Kuna so ku koma zuwa Sashe na 5 na jerin LFCS (\Yadda ake Dutsen/Unmount Local and Network (Samba & NFS) Filesystems in Linux) don cikakkun bayanai kan hawa hannun jari na nesa na NFS akan buƙata. ta amfani da umarnin Mount ko ta dindindin ta fayil ɗin /etc/fstab.

Rashin haɓaka tsarin fayil ɗin hanyar sadarwa ta amfani da waɗannan hanyoyin shine tsarin dole ne ya ware abubuwan da suka dace don kiyaye rabon a kowane lokaci, ko aƙalla har sai mun yanke shawarar cire su da hannu. Wata hanyar ita ce shigar da tsarin fayil ɗin da ake so akan buƙata ta atomatik (ba tare da yin amfani da umarnin mount ba) ta hanyar autofs, wanda zai iya hawa tsarin fayil idan an yi amfani da su sannan a cire su bayan. lokacin rashin aiki.

Autofs yana karanta /etc/auto.master, wanda ke da tsari mai zuwa:

[mount point]	[map file]

Inda aka yi amfani da [fayil ɗin taswira] don nuna maki da yawa a cikin [mount point].

Ana amfani da wannan babban fayil ɗin taswirar taswirar (/etc/auto.master) don tantance wuraren da aka fayyace, sannan a fara tsari na atomatik tare da ƙayyadaddun sigogi na kowane wurin dutsen.

Shirya /etc/auto.master kamar haka:

/media/nfs	/etc/auto.nfs-share	--timeout=60

kuma ƙirƙirar fayil ɗin taswira mai suna /etc/auto.nfs-share tare da abubuwan ciki masu zuwa:

writeable_share  -fstype=nfs4 192.168.0.10:/
non_writeable_share  -fstype=nfs4 192.168.0.10:/mydir

Lura cewa filin farko a cikin /etc/auto.nfs-share shine sunan babban kundin adireshi a cikin /media/nfs. Ana ƙirƙira kowane ƙaramin kundin adireshi da ƙarfi ta autofs.

Yanzu, sake kunna sabis na autofs:

# service autofs restart 			[sysvinit / upstart based systems]
# systemctl restart autofs 			[systemd based systems]

kuma a ƙarshe, don kunna autofs don farawa akan boot, gudanar da umarni mai zuwa:

# chkconfig --level 345 autofs on
# systemctl enable autofs 			[systemd based systems]

Lokacin da muka sake kunnawa autofs, umarnin mount yana nuna mana cewa an ɗora fayil ɗin taswirar (/etc/auto.nfs-share) akan ƙayyadaddun bayanai. directory a cikin /etc/auto.master:

Lura cewa ba a haƙiƙan kundayen adireshi da aka ɗora ba tukuna, amma za su kasance ta atomatik lokacin da muka yi ƙoƙarin samun damar hannun jari da aka kayyade a /etc/auto.nfs-share:

Kamar yadda muke iya gani, sabis ɗin autofs \yana hawa fayil ɗin taswirar, don yin magana, amma yana jira har sai an nemi tsarin fayilolin don hawa su.

Zaɓuɓɓukan annuid da anongid, tare da root_squash kamar yadda aka saita a cikin kashi na farko, suna ba mu damar yin taswirar buƙatun da tushen mai amfani ya yi a cikin abokin ciniki zuwa asusun gida a cikin uwar garken.

A wasu kalmomi, lokacin da tushen tushen abokin ciniki ya ƙirƙiri fayil a waccan kundin da aka fitar, ikon mallakarsa za a yi taswira ta atomatik zuwa asusun mai amfani tare da UID da GID = 1000, muddin akwai irin wannan asusun akan sabar:

Kammalawa

Ina fatan kun sami nasarar saitawa da saita uwar garken NFS wanda ya dace da yanayin ku ta amfani da wannan labarin azaman jagora. Hakanan kuna iya komawa zuwa shafukan mutum masu dacewa don ƙarin taimako (mai fitarwa da man idmapd.conf, misali).

Jin kyauta don gwaji tare da wasu zaɓuɓɓuka da shari'o'in gwaji kamar yadda aka zayyana a baya kuma kada ku yi jinkirin amfani da fom ɗin da ke ƙasa don aika ra'ayoyinku, shawarwari, ko tambayoyinku. Za mu yi farin cikin ji daga gare ku.