Koozali.org: home of the SME Server

Other Languages => Italiano => Topic started by: filloweb on April 01, 2010, 02:31:34 PM

Title: Comportamento anomalo MDADM
Post by: filloweb on April 01, 2010, 02:31:34 PM
Ciao a tutti.

Ho installato SME su un server con controller RAID Adaptec I2O con 4 dischi da 146SCSI in RAID 5. Da controller vedo correttamente il volume da 400GB e da sistema installato il device e' /dev/i20/hda

Purtroppo xo', mi segnala continuamente che il raid e' degradato su tutti e tre i device md ( md1, md2, md3 ).
Secondo me, xo', e' un problema software della gestione di MDADM....

Eccovi come appare configurato:

Code: [Select]
[root@mail etc]# more /etc/mdadm.conf

# mdadm.conf written out by anaconda
DEVICE partitions
ARRAY /dev/md1 level=raid1 num-devices=2 UUID=084e3408:cba1f287:eade4522:23417ef8
ARRAY /dev/md2 level=raid1 num-devices=2 UUID=18328bf8:a9650c1c:d628a8ca:55ed7c90
ARRAY /dev/md3 level=raid1 num-devices=2 UUID=33664e75:0af2b523:21022f35:0102a13e

Code: [Select]
[root@mail etc]# more /proc/mdstat
Personalities : [raid1]
md2 : active raid1 i2o/hda3[0]
      2040128 blocks [2/1] [U_]

md3 : active raid1 i2o/hda2[0]
      427312832 blocks [2/1] [U_]

md1 : active raid1 i2o/hda1[0]
      104320 blocks [2/1] [U_]

unused devices: <none>
Code: [Select]
[root@mail etc]# mdadm --detail /dev/md1
/dev/md1:
        Version : 00.90.01
  Creation Time : Wed Mar 31 13:00:14 2010
     Raid Level : raid1
     Array Size : 104320 (101.89 MiB 106.82 MB)
    Device Size : 104320 (101.89 MiB 106.82 MB)
   Raid Devices : 2
  Total Devices : 1
Preferred Minor : 1
    Persistence : Superblock is persistent

    Update Time : Thu Apr  1 11:27:32 2010
          State : clean, degraded
 Active Devices : 1
Working Devices : 1
 Failed Devices : 0
  Spare Devices : 0

           UUID : 084e3408:cba1f287:eade4522:23417ef8
         Events : 0.486

    Number   Major   Minor   RaidDevice State
       0      80        1        0      active sync   /dev/i2o/hda1
       1       0        0        -      removed
Code: [Select]
[root@mail etc]# mdadm --detail /dev/md2
/dev/md2:
        Version : 00.90.01
  Creation Time : Wed Mar 31 13:00:14 2010
     Raid Level : raid1
     Array Size : 2040128 (1992.65 MiB 2089.09 MB)
    Device Size : 2040128 (1992.65 MiB 2089.09 MB)
   Raid Devices : 2
  Total Devices : 1
Preferred Minor : 2
    Persistence : Superblock is persistent

    Update Time : Wed Mar 31 18:29:46 2010
          State : clean, degraded
 Active Devices : 1
Working Devices : 1
 Failed Devices : 0
  Spare Devices : 0

           UUID : 18328bf8:a9650c1c:d628a8ca:55ed7c90
         Events : 0.14

    Number   Major   Minor   RaidDevice State
       0      80        3        0      active sync   /dev/i2o/hda3
       1       0        0        -      removed
Code: [Select]
[root@mail etc]# mdadm --detail /dev/md3
/dev/md3:
        Version : 00.90.01
  Creation Time : Wed Mar 31 13:00:14 2010
     Raid Level : raid1
     Array Size : 427312832 (407.52 GiB 437.57 GB)
    Device Size : 427312832 (407.52 GiB 437.57 GB)
   Raid Devices : 2
  Total Devices : 1
Preferred Minor : 3
    Persistence : Superblock is persistent

    Update Time : Thu Apr  1 14:26:19 2010
          State : clean, degraded
 Active Devices : 1
Working Devices : 1
 Failed Devices : 0
  Spare Devices : 0

           UUID : 33664e75:0af2b523:21022f35:0102a13e
         Events : 0.27174

    Number   Major   Minor   RaidDevice State
       0      80        2        0      active sync   /dev/i2o/hda2
       1       0        0        -      removed

Secondo voi e' sufficente modificare /etc/mdadm.conf dichiarando num-devices=1 trasformandolo cosi:
Code: [Select]
[root@mail etc]# more /etc/mdadm.conf

# mdadm.conf written out by anaconda
DEVICE partitions
ARRAY /dev/md1 level=raid1 num-devices=1 UUID=084e3408:cba1f287:eade4522:23417ef8
ARRAY /dev/md2 level=raid1 num-devices=1 UUID=18328bf8:a9650c1c:d628a8ca:55ed7c90
ARRAY /dev/md3 level=raid1 num-devices=1 UUID=33664e75:0af2b523:21022f35:0102a13e
Grazie per l'aiuto.
filloweb

 
Title: Re: Comportamento anomalo MDADM
Post by: nicolatiana on April 01, 2010, 02:57:51 PM
Quote
Da controller vedo correttamente il volume da 400GB
Da quanto sopra si deduce che hai configurato il raid 5 sul controller? O no ?
Se si le informazioni di mdam non sono significative. Se vuoi usare mdam (raid sw) devi lasciare i dischi collegati al controller come dischi singoli e POI installare SME usando la cfg raid di serie (RAID 5+hs).
Verificare
Saluti
Nicola
Title: Re: Comportamento anomalo MDADM
Post by: filloweb on April 01, 2010, 03:12:58 PM
Infatti.

Ho un RAID hardware gestito dal controller.

SME non dovrebbe pero' darmi la segnalazione di RAID degradato... Dovrebbe solo avere dei device md con un'unica partizione attiva e configurata... e non una attiva e una degradata...

Qualcosa non mi torna...
Se faccio una installazione su disco singolo, non mi segnala continuamente RAID degradato...

Non capisco...
filloweb
Title: Re: Comportamento anomalo MDADM
Post by: nicolatiana on April 01, 2010, 03:40:24 PM
L'unico strumento che può monitorare correttamente il tuo raid è (se esiste) un tool adaptec per REHL4.
Per mdadm il tuo disco non vuol dire nulla, anzi il fatto che sia un controller RAID contribuisce a ingarbugliare la faccenda.
Saluti
Nicola
Title: Re: Comportamento anomalo MDADM
Post by: filloweb on April 01, 2010, 03:57:57 PM
Scusa ma allora NON mi sono spiegato.

Non me ne frega di monitorarlo, mi interessa solo che MDADM NON mi segnali un errore nel RAID quando in realta' e' tutto OK.

Grazie.
filloweb
Title: Re: Comportamento anomalo MDADM
Post by: Fumetto on April 01, 2010, 04:05:30 PM
Forse nicola intendeva dire che MDADM si "incacchia" perchè il raid non lo gestisce lui e quindi ti da gli errori per questo motivo...
Title: Re: Comportamento anomalo MDADM
Post by: nicolatiana on April 01, 2010, 04:34:47 PM
Se mdadm esiste e funziona cerca di fare il suo lavoro; è possibile che il tuo hw gestisca i dischi in un modo abbastanza simile a mdam e questo crea confusione.

Per eliminare il problema alla radice puoi installare SME senza il supporto Raid
To install SME without software RAID type:
Quote
sme noraid
Quanto a non installare i tool di monitoraggio vedi tu: senza non hai nessuno strumento che ti avvisi (come fa mdam con le mail a admin) a parte i led gialli dei dischi HS o i fischi del controller.
Saluti
Nicola

Title: Re: Comportamento anomalo MDADM
Post by: Stefano on April 01, 2010, 11:16:22 PM
Quote from: filloweb on April 01, 2010, 03:57:57 PM
Scusa ma allora NON mi sono spiegato.

Non me ne frega di monitorarlo, mi interessa solo che MDADM NON mi segnali un errore nel RAID quando in realta' e' tutto OK.

Grazie.
filloweb


Code: [Select]
config setprop raidmonitor status disabled
service raidmonitor stop

HTH

p.s. sto ancora aspettando tua risposta ;-)
Title: Re: Comportamento anomalo MDADM
Post by: CharlieBrady on April 09, 2010, 10:03:28 PM
Quote from: Stefano on April 01, 2010, 11:16:22 PM
Code: [Select]
config setprop raidmonitor status disabled
service raidmonitor stop

The above should not be necessary. The raidmonitor run script checks for the number of disks in the system. If there is only one "disk" (or more than one disk, but with hardware RAID), the run script shuts down the raidmonitor service.

Does the OP actually have a problem? Or is he/she just worried about what appears when "mdadm --detail" is run?
Title: Re: Comportamento anomalo MDADM
Post by: Stefano on April 09, 2010, 10:12:08 PM
Hi Charlie..

OP is getting (was, we haven't had any feedback from him) mail form mdadm monitoring saying that raid is working in degraded mode, and he wish to get rid of the annoying mails
Title: Re: Comportamento anomalo MDADM
Post by: CharlieBrady on April 09, 2010, 11:09:52 PM
Quote from: Stefano on April 09, 2010, 10:12:08 PM
OP is getting (was, we haven't had any feedback from him) mail form mdadm monitoring saying that raid is working in degraded mode, and he wish to get rid of the annoying mails

OP shouldn't have been getting those mails. Bug report, with details, please. :-)
Title: Re: Comportamento anomalo MDADM
Post by: CharlieBrady on April 09, 2010, 11:14:55 PM
Quote from: Stefano on April 09, 2010, 10:12:08 PM
Hi Charlie..

OP is getting (was, we haven't had any feedback from him) mail form mdadm monitoring saying that raid is working in degraded mode, and he wish to get rid of the annoying mails

Google's translation doesn't say anything about email. It just says:

 Unfortunately xo ', constantly tells me that the raid and' degraded all three devices md (MD1, MD2, MD3).
 In my opinion, x ', is a problem management software .... mdadm

Does "xo'" signify email?

I still think that OP might be (mis-)interpreting the output of mdadm.

 
Title: Re: Comportamento anomalo MDADM
Post by: Stefano on April 09, 2010, 11:38:57 PM
Quote from: CharlieBrady on April 09, 2010, 11:09:52 PM
OP shouldn't have been getting those mails. Bug report, with details, please. :-)

I hope he will read your request :-)
Title: Re: Comportamento anomalo MDADM
Post by: Stefano on April 09, 2010, 11:40:35 PM
Quote from: CharlieBrady on April 09, 2010, 11:14:55 PM
Does "xo'" signify email?

xo' is a orrible contraction of "però" that you can translate with "but".. you can safely ignore it :-)
Title: Re: Comportamento anomalo MDADM
Post by: CharlieBrady on April 10, 2010, 02:16:03 AM
In that case I see no evidence that email was sent. I think OP was running mdadm, and mis-interpreting the output (which is normal).
Title: Re: Comportamento anomalo MDADM
Post by: Stefano on April 10, 2010, 10:38:26 AM
Hi Charlie..

we are all waiting for OP answer.. I will let you know.. thank you