<div dir="ltr">The import is showing the following though the OST is formatted with <span style="font-family:monospace">failover.node=10.99.100.152 or service mode </span><span style="font-family:monospace">=10.99.100.152</span><div><br></div><div><span style="font-family:monospace">    connection:</span><br style="font-family:monospace"><span style="font-family:monospace">       failover_nids: [ 0@lo, 0@lo ]</span><br style="font-family:monospace"><span style="font-family:monospace">       current_connection: 0@lo</span></div><div><span style="font-family:monospace"><br></span></div><div><span style="font-family:monospace"><br></span></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, 5 Nov 2024 at 12:34, Backer <<a href="mailto:backer.kolo@gmail.com">backer.kolo@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">Hi,<div><br></div>Mounting the Lustre file file system on the OSS. Some of the OSTs are locally attached to the OSS. <br><br>The failover IP on the OST is "10.99.100.152". It is a local lnet on the OSS. However, when the client mounts it, the import automatically changes to 0@lo. It is undesirable here because when this OST fails over to another server, the client is still trying to connect to 0@lo while it is no longer on the same host. This makes the client fs mount hangs for ever.  <br><br>Here the failover is designed in such a way that the IP address moves (fails over) with OST and becomes active on the other server. <br><br>How can I make the import pointing to the real IP and not the loopback? (so that the failover works)<div><span style="font-family:monospace"></span></div><div><span style="font-family:monospace"><br></span></div><div><br></div><font face="monospace">[oss000 ~]$ lfs df<br>UUID                   1K-blocks        Used   Available Use% Mounted on<br>fs-MDT0000_UUID     29068444       25692    26422344   1% /mnt/fs[MDT:0]<br>fs-OST0000_UUID     50541812    30160292    17743696  63% /mnt/fs[OST:0]<br>fs-OST0001_UUID     50541812    29301740    18602248  62% /mnt/fs[OST:1]<br>fs-OST0002_UUID     50541812    29356508    18547480  62% /mnt/fs[OST:2]<br>fs-OST0003_UUID     50541812     8822980    39081008  19% /mnt/fs[OST:3]<br><br>filesystem_summary:    202167248    97641520    93974432  51% /mnt/fs<br><br>[oss000 ~]$ df -h<br>Filesystem                  Size  Used Avail Use% Mounted on<br>devtmpfs                     30G     0   30G   0% /dev<br>tmpfs                        30G  8.1M   30G   1% /dev/shm<br>tmpfs                        30G   25M   30G   1% /run<br>tmpfs                        30G     0   30G   0% /sys/fs/cgroup<br>/dev/mapper/ocivolume-root   36G   17G   19G  48% /<br>/dev/sdc2                  1014M  637M  378M  63% /boot<br>/dev/mapper/ocivolume-oled   10G  2.5G  7.6G  25% /var/oled<br>/dev/sdc1                   100M  5.1M   95M   6% /boot/efi<br>tmpfs                       5.9G     0  5.9G   0% /run/user/987<br>tmpfs                       5.9G     0  5.9G   0% /run/user/0<br>/dev/sdb                     49G   28G   18G  62% /fs-OST0001<br>/dev/sda                     49G   29G   17G  63% /fs-OST0000<br>tmpfs                       5.9G     0  5.9G   0% /run/user/1000<br>10.99.100.221@tcp1:/fs  193G   94G   90G  51% /mnt/fs<br><br>[oss000 ~]$ sudo tunefs.lustre --dryrun /dev/sda<br>checking for existing Lustre data: found<br><br>   Read previous values:<br>Target:     fs-OST0000<br>Index:      0<br>Lustre FS:  fs<br>Mount type: ldiskfs<br>Flags:      0x1002<br>              (OST no_primnode )<br>Persistent mount opts: ,errors=remount-ro<br>Parameters: mgsnode=10.99.100.221@tcp1 failover.node=10.99.100.152@tcp1,10.99.100.152@tcp1<br><br><br>   Permanent disk data:<br>Target:     fs-OST0000<br>Index:      0<br>Lustre FS:  fs<br>Mount type: ldiskfs<br>Flags:      0x1002<br>              (OST no_primnode )<br>Persistent mount opts: ,errors=remount-ro<br>Parameters: mgsnode=10.99.100.221@tcp1 failover.node=10.99.100.152@tcp1,10.99.100.152@tcp1<br><br>exiting before disk write.<br><br><br>[oss000 proc]# cat /proc/fs/lustre/osc/fs-OST0000-osc-ffff89c57672e000/import<br>import:<br>    name: fs-OST0000-osc-ffff89c57672e000<br>    target: fs-OST0000_UUID<br>    state: IDLE<br>    connect_flags: [ write_grant, server_lock, version, request_portal, max_byte_per_rpc, early_lock_cancel, adaptive_timeouts, lru_resize, alt_checksum_algorithm, fid_is_enabled, version_recovery, grant_shrink, full20, layout_lock, 64bithash, object_max_bytes, jobstats, einprogress, grant_param, lvb_type, short_io, lfsck, bulk_mbits, second_flags, lockaheadv2, increasing_xid, client_encryption, lseek, reply_mbits ]<br>    connect_data:<br>       flags: 0xa0425af2e3440078<br>       instance: 39<br>       target_version: 2.15.3.0<br>       initial_grant: 8437760<br>       max_brw_size: 4194304<br>       grant_block_size: 4096<br>       grant_inode_size: 32<br>       grant_max_extent_size: 67108864<br>       grant_extent_tax: 24576<br>       cksum_types: 0xf7<br>       max_object_bytes: 17592186040320<br>    import_flags: [ replayable, pingable, connect_tried ]<br>    connection:<br>       failover_nids: [ 0@lo, 0@lo ]<br>       current_connection: 0@lo<br>       connection_attempts: 1<br>       generation: 1<br>       in-progress_invalidations: 0<br>       idle: 36 sec<br>    rpcs:<br>       inflight: 0<br>       unregistering: 0<br>       timeouts: 0<br>       avg_waittime: 2627 usec<br>    service_estimates:<br>       services: 1 sec<br>       network: 1 sec<br>    transactions:<br>       last_replay: 0<br>       peer_committed: 0<br>       last_checked: 0</font></div>
</blockquote></div>