|
|
Sponsored Content
Operating Systems
Solaris
Solstice Disksuite Question
Post 302086545 by BOFH on Thursday 24th of August 2006 08:47:22 PM
08-24-2006
Solstice Disksuite Question
According to the metadb man page:
Now based on italicized part, if a mirror only has a single replica and never crashes, it should still function.
Same with booting. I thought three replicas were necessary for the system to be able to come up but the comment about crashing tells me that as long as the mirrors are stable (iow never crash), the system should come up.
The reason I ask this is that I've started a new position and discovered 8 disks in need of maintenance (metareplace), one disk out of sync (all partitions) and one disk with the status of Unknown.
In addition, I found several replicas that have failed (W - write errors and M - replica had problem with master blocks) and quite a few with less than 3 replicas on a disk (I prefer to configure a system with a mininum of 3 replicas per disk since disks are so large). At least one system had only a single replica database and 2 or 3 had two replica databases.
Per the comment in the man page, you need a majority (technically one is a majority) for the state information to be validated.
The problem is that systems are functioning correctly with one or two replicas and the one with a single replica was rebooted four days ago. Since it was booted, I have to believe that three replicas aren't required. In fact, as long as the replica db is stable, you can boot with a single replica (based on evidence).
So:
How does a system with only two replicas still manage to operate correctly? I can logically see how one will work (majority) but only two should fail.
Are my thoughts on a single replica and the system not crashing accurate? Or at least logical
My next steps appear to be to get the bad disks replaced (easy enough) but also to get the replicas in order. For the systems with 1 or 2 replicas, I think I'm going to have to break the mirrors and rebuild with what I think is the correct number of replicas (some of the disks have 5 replicas per disk).
Thoughts? Pointers to more technical information that what's on docs.sun.com?
Thanks.
Carl
Quote:
Replicated databases have an inherent problem in determining
which database has valid and correct data. To solve this
problem, Volume Manager uses a majority consensus algorithm.
This algorithm requires that a majority of the database
replicas be available before any of them are declared valid.
This algorithm strongly encourages the presence of at least
three initial replicas, which you create. A consensus can
then be reached as long as at least two of the three repli-
cas are available. If there is only one replica and the sys-
tem crashes, it is possible that all metadevice configura-
tion data can be lost.
which database has valid and correct data. To solve this
problem, Volume Manager uses a majority consensus algorithm.
This algorithm requires that a majority of the database
replicas be available before any of them are declared valid.
This algorithm strongly encourages the presence of at least
three initial replicas, which you create. A consensus can
then be reached as long as at least two of the three repli-
cas are available. If there is only one replica and the sys-
tem crashes, it is possible that all metadevice configura-
tion data can be lost.
Same with booting. I thought three replicas were necessary for the system to be able to come up but the comment about crashing tells me that as long as the mirrors are stable (iow never crash), the system should come up.
The reason I ask this is that I've started a new position and discovered 8 disks in need of maintenance (metareplace), one disk out of sync (all partitions) and one disk with the status of Unknown.
In addition, I found several replicas that have failed (W - write errors and M - replica had problem with master blocks) and quite a few with less than 3 replicas on a disk (I prefer to configure a system with a mininum of 3 replicas per disk since disks are so large). At least one system had only a single replica database and 2 or 3 had two replica databases.
Per the comment in the man page, you need a majority (technically one is a majority) for the state information to be validated.
The problem is that systems are functioning correctly with one or two replicas and the one with a single replica was rebooted four days ago. Since it was booted, I have to believe that three replicas aren't required. In fact, as long as the replica db is stable, you can boot with a single replica (based on evidence).
So:
How does a system with only two replicas still manage to operate correctly? I can logically see how one will work (majority) but only two should fail.
Are my thoughts on a single replica and the system not crashing accurate? Or at least logical
My next steps appear to be to get the bad disks replaced (easy enough) but also to get the replicas in order. For the systems with 1 or 2 replicas, I think I'm going to have to break the mirrors and rebuild with what I think is the correct number of replicas (some of the disks have 5 replicas per disk).
Thoughts? Pointers to more technical information that what's on docs.sun.com?
Thanks.
Carl
|
|
10 More Discussions You Might Find Interesting
1. UNIX for Dummies Questions & Answers
Dear all
i search a linear RAID Howto for the solaris os. Wiht the help of google and docs.sunc.com i found nothing...
can you share your knowlage wiht me?
thanx
adme (1 Reply)
Discussion started by: adme
1 Replies
2. UNIX for Dummies Questions & Answers
hi, i tried these sun command on a sun0s 5.7 but get the "cant open display" message. please advise. i am using netterm to telnet. (5 Replies)
Discussion started by: yls177
5 Replies
3. UNIX for Dummies Questions & Answers
I am brand new to UNIX and have been given the task to remove veritas volume manager 3.5 mirroring and install Disksuite mirror on two Solaris 5.8 servers. Does anyone know where I can find step by step instructions to perform these tasks?
Thanks (1 Reply)
Discussion started by: mg2
1 Replies
4. Solaris
Has anybody every used Solstice DiskSuite? I am having trouble setting it up. I installed it without a problem, but do I really have to blow away the drives on the D1000 just to create a metastate database? (8 Replies)
Discussion started by: hshapiro
8 Replies
5. Solaris
I'm trying to figure out how to simply create a 500Gb ufs file system named
/rec using DiskSuite 4.2.1 on a Solaris 8 platform via command line using two
250Gb partitions on two different disks. Does anyone know what the proper
command string and options are to perform this task? Would I use the... (2 Replies)
Discussion started by: ruger14
2 Replies
6. Solaris
Hi all,
Do you know where I can download Soltice Disksuite 4.2 for Solaris 2.6 ?
I haven't the CD labeled “Solaris Server Intranet Extensions 1.0” .
Thanks in advance for your precious help.
Bests Regards
Hosni (2 Replies)
Discussion started by: hosni
2 Replies
7. Solaris
Hi people,
I have a Solstice Backup 6.0.3 installed in server X1 and one installed in server X2.
I need to start de recover in X2 and the somes files existing in X1.
- Well, i begin this process in X2 with
#nwadmin -s X1 &
and i select "Save Set/ Recover". I select the datas and click in... (0 Replies)
Discussion started by: denisgomes
0 Replies
8. Solaris
I am doing disk mirroring and this command I entered:
metainit d202 1 1 c0t1d0s0
gives me this error "metainit: d202: no such file or directory"
The nmd value in /kernel/drv/md.conf is set to 2000.
Any ideas what this error means? (3 Replies)
Discussion started by: bluridge
3 Replies
9. Solaris
Hello all,
I have a Solaris Disksuite question :-
I will be adding 4 new drives to an E250 server and need will be configuring 2 striped volumes each consisting 2 new disks with SVM. In the end i will have 2 volumes each of 72gb. So in effect i will have 1 volume called D7 and another volume... (6 Replies)
Discussion started by: commandline
6 Replies
10. Solaris
Hi
I have an existing mirror configured (d10) with submirrors (d11 and d12). I recently detached the d12 mirror for an upgrade. I know want to re-attach the d12 mirror to sync with the active mirror (d11). When I run the command metattach d10 d12 - which way will the sync occur?
Will the... (1 Reply)
Discussion started by: samruthroy
1 Replies
LEARN ABOUT DEBIAN
gfrep
GFREP(1) GFREP(1) NAME
gfrep - replicate Gfarm files SYNOPSIS
gfrep [ -mnqvx ] [ -S source-domainname ] [ -D destination-domainname ] [ -h source-hostfile ] [ -H destination-hostfile ] [ -N num-of- replicas ] [ -j num-of-threads ] path ... DESCRIPTION
gfrep creates specified number of replicas of Gfarm files specified by the path to specified set of destination hosts. When set of source hosts are specified, only files stored on the set of source hosts are replicated. A set of hosts can be specified by a domain name and/or a host file. When both a domain name and a host file are specified, a set of hosts is determined by both conditions. If the path parameter is a directory, files under the directory will be copied recursively. OPTIONS
-S source-domainname Specifies the domain name of source filesystem nodes from which replicas will be copied. -D destination-domainname Specifies the domain name of destination filesystem nodes to which replicas will be copied. If neither this nor the -H option is specified, replicas may be copied to any available host. -h source-hostfile Specifies a file which describes hostnames of source filesystem nodes from which the replica will be made. The source-hostfile con- sists of a filesystem node name on each line. if ``-'' is specified, standard input is used to read the host list. -H destination-hostfile Specifies a file which describes hostnames of destination filesystem nodes on which the replica will be made. The destination-host- file consists of a filesystem node name on each line. if ``-'' is specified, standard input is used to read the host list. -N num-of-replicas Specifies the required number of file replicas, counting already existing ones. If there are a sufficient number of replicas in the destination hosts, neither any copying nor any removal will be done. If this option is not specified, -N 1 is assumed. -j num-of-threads Specifies the number of threads to create file replicas simultaneously. This option is only available when compiled by an OpenMP C compiler. By default, it will be the smallest among the number of total files, the number of source hosts, and the number of desti- nation hosts. -m Lets gfrep command do a migrate/move, instead of doing replication. -x Removes file replicas when a set of destination nodes has more file replicas than specified. -n Displays files to be replicated, but does nothing. -q Suppresses non-error messages. -v Displays verbose output. -? Displays a list of command options. BUGS
A source host is not always selected within a set of source hosts. Gfarm 09 November 2009 GFREP(1)