Esercitazione 3

Subnet

Controlla le interfacce di rete e verifica connettivita' su eth1 con gli altri nodi del cluster

<html> # ip addr show 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:16:3e:02:01:2e brd ff:ff:ff:ff:ff:ff inet 192.168.200.246/24 brd 192.168.200.255 scope global eth0 inet6 fe80::216:3eff:fe02:12e/64 scope link valid_lft forever preferred_lft forever 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 52:54:00:9d:5b:c5 brd ff:ff:ff:ff:ff:ff inet 10.168.200.246/24 brd 10.168.200.255 scope global eth1 inet6 fe80::5054:ff:fe9d:5bc5/64 scope link valid_lft forever preferred_lft forever # ping -c 2 10.168.200.247 PING 10.168.200.247 (10.168.200.247) 56(84) bytes of data. 64 bytes from 10.168.200.247: icmp_seq=1 ttl=64 time=1.77 ms 64 bytes from 10.168.200.247: icmp_seq=2 ttl=64 time=0.279 ms — 10.168.200.247 ping statistics — 2 packets transmitted, 2 received, 0% packet loss, time 1002ms rtt min/avg/max/mdev = 0.279/1.027/1.775/0.748 ms # ping -c 2 10.168.200.248 PING 10.168.200.248 (10.168.200.248) 56(84) bytes of data. 64 bytes from 10.168.200.248: icmp_seq=1 ttl=64 time=1.02 ms 64 bytes from 10.168.200.248: icmp_seq=2 ttl=64 time=0.215 ms — 10.168.200.248 ping statistics — 2 packets transmitted, 2 received, 0% packet loss, time 1001ms rtt min/avg/max/mdev = 0.215/0.619/1.024/0.405 ms </html>

Definisce la rete IP delle interfacce eth1 come preferenziale per il cluster

<html> # mmchconfig subnets="10.168.200.0" mmchconfig: Command successfully completed mmchconfig: Propagating the cluster configuration data to all affected nodes. This is an asynchronous process. </html>

Verifica la attuale configurazione della connettivita' del nodo su cui operi

# mmdiag –network === mmdiag: network === Pending messages: (none) Inter-node communication configuration: tscTcpPort 1191 my address 192.168.200.246/24 (eth0) <c0n0> my addr list 192.168.200.246/24 (eth0) my node number 1 TCP Connections between nodes: Device eth0: hostname node destination status err sock sent(MB) recvd(MB) ostype c16-serv2 <c0n1> 192.168.200.247 connected 0 15 0 0 Linux/L c16-serv3 <c0n2> 192.168.200.248 connected 0 23 0 0 Linux/L diag verbs: VERBS RDMA class not initialized </html>

Restart dei demoni e verifica della connettivita' dopo il boot

<html> # mmshutdown -a Wed Nov 26 13:30:22 CET 2014: mmshutdown: Starting force unmount of GPFS file systems Wed Nov 26 13:30:27 CET 2014: mmshutdown: Shutting down GPFS daemons c16-serv1.openlab.infn.it: Shutting down! c16-serv3.openlab.infn.it: Shutting down! c16-serv2.openlab.infn.it: Shutting down! c16-serv1.openlab.infn.it: 'shutdown' command about to kill process 29670 c16-serv1.openlab.infn.it: Unloading modules from /lib/modules/2.6.32-358.el6.x86_64/extra c16-serv3.openlab.infn.it: 'shutdown' command about to kill process 20207 c16-serv3.openlab.infn.it: Unloading modules from /lib/modules/2.6.32-358.el6.x86_64/extra c16-serv1.openlab.infn.it: Unloading module mmfs26 c16-serv3.openlab.infn.it: Unloading module mmfs26 c16-serv2.openlab.infn.it: 'shutdown' command about to kill process 15607 c16-serv2.openlab.infn.it: Unloading modules from /lib/modules/2.6.32-358.el6.x86_64/extra c16-serv2.openlab.infn.it: Unloading module mmfs26 c16-serv3.openlab.infn.it: Unloading module mmfslinux c16-serv3.openlab.infn.it: Unloading module tracedev c16-serv1.openlab.infn.it: Unloading module mmfslinux c16-serv1.openlab.infn.it: Unloading module tracedev c16-serv2.openlab.infn.it: Unloading module mmfslinux c16-serv2.openlab.infn.it: Unloading module tracedev Wed Nov 26 13:30:35 CET 2014: mmshutdown: Finished # mmstartup -a Wed Nov 26 13:30:39 CET 2014: mmstartup: Starting GPFS … # mmdiag –network === mmdiag: network === Pending messages: (none) Inter-node communication configuration: tscTcpPort 1191 my address 192.168.200.246/24 (eth0) <c0n0> my addr list 10.168.200.246/24 (eth1) 192.168.200.246/24 (eth0) my node number 1 TCP Connections between nodes: Device eth1: hostname node destination status err sock sent(MB) recvd(MB) ostype c16-serv2 <c0n1> 10.168.200.247 connected 0 21 0 0 Linux/L c16-serv3 <c0n2> 10.168.200.248 connected 0 15 0 0 Linux/L diag verbs: VERBS RDMA class not initialized </html>

Snapshot

Crea una snapshot globale sul file system c16fs

<html> # dd if=/dev/zero of=/gpfs/c16fs/datafile bs=1M count=1K 1024+0 records in 1024+0 records out 1073741824 bytes (1.1 GB) copied, 5.33226 s, 201 MB/s # dd if=/dev/zero of=/gpfs/c16fs/fset1/datafile bs=1M count=1K 1024+0 records in 1024+0 records out 1073741824 bytes (1.1 GB) copied, 5.41672 s, 198 MB/s # dd if=/dev/zero of=/gpfs/c16fs/fset2/datafile bs=1M count=1K 1024+0 records in 1024+0 records out 1073741824 bytes (1.1 GB) copied, 5.55623 s, 193 MB/s # mkdir /gpfs/c16fs/testdir # mkdir /gpfs/c16fs/fset1/testdir # mkdir /gpfs/c16fs/fset2/testdir # df -k /gpfs/c16fs/ /gpfs/c16fs/fset1 /gpfs/c16fs/fset2 Filesystem 1K-blocks Used Available Use% Mounted on /dev/c16fs 4194304000 3695616 4190608384 1% /gpfs/c16fs /dev/c16fs 10485760 1048576 9437184 10% /gpfs/c16fs /dev/c16fs 10485760 1048576 9437184 10% /gpfs/c16fs # mmcrsnapshot c16fs global_snap1 Writing dirty data to disk Quiescing all file system operations Writing dirty data to disk again Resuming operations. Snapshot global_snap1 created with id 5. </html>

Visualizza la snapshot (sia l'occupazione che l'accesso al contenuto)

<html> # mmlssnapshot c16fs -d Snapshots in file system c16fs: [data and metadata in KB] Directory SnapId Status Created Fileset Data Metadata global_snap1 5 Valid Wed Nov 26 16:10:30 2014 0 1024 # ls /gpfs/c16fs/.snapshots/ global_snap1 </html>

Abilita la visualizzazione delle snapshot all'interno di ogni directory

<html> # ls /gpfs/c16fs/testdir/.snapshots/ ls: cannot access /gpfs/c16fs/testdir/.snapshots/: No such file or directory # mmsnapdir /dev/c16fs -a # ls /gpfs/c16fs/testdir/.snapshots/ global_snap1 </html>

Verifica occupazione in occasione di modifiche

<html> # ls -l /gpfs/c16fs/datafile -rw-r–r– 1 root root 1073741824 Nov 26 13:51 /gpfs/c16fs/datafile # mmlssnapshot /dev/c16fs -d Snapshots in file system c16fs: [data and metadata in KB] Directory SnapId Status Created Fileset Data Metadata global_snap1 5 Valid Wed Nov 26 16:10:30 2014 0 1024 # rm -f /gpfs/c16fs/datafile # mmlssnapshot /dev/c16fs -d Snapshots in file system c16fs: [data and metadata in KB] Directory SnapId Status Created Fileset Data Metadata global_snap1 5 Valid Wed Nov 26 16:10:30 2014 1048608 2080 </html>

Crea sanpshot su independent fileset e verifica che la visibilita' e' solo all'interno dello stesso

<html> # mmcrsnapshot /dev/c16fs snap2 -j fset2 Writing dirty data to disk Quiescing all file system operations Writing dirty data to disk again Resuming operations. Snapshot snap2 created with id 6. # mmlssnapshot /dev/c16fs -d Snapshots in file system c16fs: [data and metadata in KB] Directory SnapId Status Created Fileset Data Metadata global_snap1 5 Valid Wed Nov 26 16:10:30 2014 1048608 2080 snap2 6 Valid Wed Nov 26 16:26:49 2014 fset2 0 0 # ls /gpfs/c16fs/.snapshots global_snap1 # ls /gpfs/c16fs/fset2/.snapshots snap2 </html>

Cloni

Creazione di un clone

<html> # cd /gpfs/c16fs/ # mkdir cloni # cd cloni # dd if=/dev/zero of=./file.orig bs=1M count=1K 1024+0 records in 1024+0 records out 1073741824 bytes (1.1 GB) copied, 6.87812 s, 156 MB/s # ls -lis total 1048576 244741 1048576 -rw-r–r– 1 root root 1073741824 Nov 27 07:05 file.orig # mmclone snap file.orig file.parent_clone # ls -lis total 1048576 244741 0 -rw-r–r– 1 root root 1073741824 Nov 27 07:05 file.orig 244742 1048576 -rw-r–r– 2 root root 1073741824 Nov 27 07:06 file.parent_clone # mmclone show * Parent Depth Parent inode File name —— —– ————– ——— no 1 244742 file.orig yes 0 file.parent_clone # mmclone copy file.parent_clone file.clone1 # ls -ils total 1048576 244743 0 -rw-r–r– 1 root root 1073741824 Nov 27 07:08 file.clone1 244741 0 -rw-r–r– 1 root root 1073741824 Nov 27 07:05 file.orig 244742 1048576 -rw-r–r– 3 root root 1073741824 Nov 27 07:06 file.parent_clone # mmclone show * Parent Depth Parent inode File name —— —– ————– ——— no 1 244742 file.clone1 no 1 244742 file.orig yes 0 file.parent_clone # dd if=/dev/zero of=./file.clone1 bs=1M count=10 10+0 records in 10+0 records out 10485760 bytes (10 MB) copied, 0.00578463 s, 1.8 GB/s # ls -ils total 1058816 244743 10240 -rw-r–r– 1 root root 10485760 Nov 27 07:09 file.clone1 244741 0 -rw-r–r– 1 root root 1073741824 Nov 27 07:05 file.orig 244742 1048576 -rw-r–r– 3 root root 1073741824 Nov 27 07:06 file.parent_clone </html>

Creazione di 100 cloni di un file da 1 GB

<html> # time for i in `seq 2 100`; do mmclone copy file.parent_clone file.clone${i}; done real 0m0.233s user 0m0.005s sys 0m0.014s # mmclone show * Parent Depth Parent inode File name —— —– ————– ——— no 1 244742 file.clone1 no 1 244742 file.clone10 no 1 244742 file.clone100 no 1 244742 file.clone11 … no 1 244742 file.clone97 no 1 244742 file.clone98 no 1 244742 file.clone99 no 1 244742 file.orig yes 0 file.parent_clone # du -sk ../cloni 1058848 ../cloni </html>

Crea gerarchia di cloni

<html> # mmclone snap file.clone1 file.parent_clone1 [root@c16-serv1 cloni]# mmcopy file.parent_clone1 file2 # mmclone copy file.parent_clone1 file2 # mmclone copy file.parent_clone1 file3 # ls -lis file.parent_clone file.parent_clone1 file.clone1 file2 file3 244843 0 -rw-r–r– 1 root root 10485760 Nov 27 07:14 file2 244844 0 -rw-r–r– 1 root root 10485760 Nov 27 07:16 file3 244743 0 -rw-r–r– 1 root root 10485760 Nov 27 07:09 file.clone1 244742 1048576 -rw-r–r– 102 root root 1073741824 Nov 27 07:06 file.parent_clone 244842 10240 -rw-r–r– 4 root root 10485760 Nov 27 07:13 file.parent_clone1 # mmclone show file.parent_clone file.parent_clone1 file.clone1 file2 file3 Parent Depth Parent inode File name —— —– ————– ——— yes 0 file.parent_clone yes 1 244742 file.parent_clone1 no 2 244842 file.clone1 no 2 244842 file2 no 2 244842 file3 </html>

Esegui uno split di file2 (diviene file ordinario) ed un redirect di file1 (diviene clone del parent dell'attuale parent). Verifica.

<html> # mmclone split file2 [root@c16-serv1 cloni]# mmclone redirect file3 [root@c16-serv1 cloni]# mmclone show file.parent_clone file.parent_clone1 file.clone1 file2 file3 Parent Depth Parent inode File name —— —– ————– ——— yes 0 file.parent_clone yes 1 244742 file.parent_clone1 no 2 244842 file.clone1 file2 no 1 244742 file3 </html>

Rimuovi i cloni

<html> # rm -f file.parent_clone rm: cannot remove `file.parent_clone': Read-only file system # rm -f file.clone? file.clone?? file.clone??? file2 # mmclone show * Parent Depth Parent inode File name —— —– ————– ——— file2 no 1 244742 file3 no 1 244742 file.orig yes 0 file.parent_clone yes 1 244742 file.parent_clone1 # rm -f file.parent_clone1 # rm -f file3 file.orig # rm -f file.parent_clone </html>

INFN wiki

Table of Contents

Esercitazione 3

Subnet

Snapshot

Cloni