Os comandos de montagem do NFS travam no cliente indefinidamente após a reinicialização do servidor.
Estou executando este comando: mount -t nfs -v 10.17.25.27:/DATA/projects/assets-db/log-apache test
Meu servidor NFS é: 10.17.25.27 , o IP do cliente é 10.17.25.152
Eu habilitei o debug do syslog para o nfs e vejo estes erros:
Aug 31 16:26:06 SEFRAPP05155 kernel: NFS call setclientid_confirm auth=UNIX, (client ID 9ee5cc5a6d8c0000)
Aug 31 16:26:06 SEFRAPP05155 kernel: NFS reply setclientid_confirm: -10008
Aug 31 16:26:06 SEFRAPP05155 kernel: NFS call setclientid auth=UNIX, 'Linux NFSv4.0 10.17.25.152/10.17.25.27 tcp'
Aug 31 16:26:06 SEFRAPP05155 kernel: NFS reply setclientid: 0
Aug 31 16:26:06 SEFRAPP05155 kernel: NFS call setclientid_confirm auth=UNIX, (client ID 9ee5cc5a6e8c0000)
Aug 31 16:26:06 SEFRAPP05155 kernel: NFS reply setclientid_confirm: -10008
Aug 31 16:26:06 SEFRAPP05155 kernel: --> nfs_put_client({8})
Aug 31 16:26:06 SEFRAPP05155 kernel: NFS: <-- nfs40_walk_client_list status = -10008
Aug 31 16:26:06 SEFRAPP05155 kernel: --> nfs4_create_server()
Aug 31 16:26:06 SEFRAPP05155 kernel: --> nfs4_init_server()
Aug 31 16:26:06 SEFRAPP05155 kernel: --> nfs4_set_client()
Aug 31 16:26:06 SEFRAPP05155 kernel: --> nfs_get_client(10.17.25.27,v4)
Aug 31 16:26:06 SEFRAPP05155 kernel: <-- nfs_found_client found nfs_client ffff8801e93a0400 for 10.17.25.27
Aug 31 16:26:06 SEFRAPP05155 kernel: <-- nfs4_set_client() = 0 [new ffff8801e93a0400]
Aug 31 16:26:06 SEFRAPP05155 kernel: <-- nfs4_init_server() = 0
Aug 31 16:26:06 SEFRAPP05155 kernel: --> nfs4_get_rootfh()
...
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS call setclientid_confirm auth=UNIX, (client ID 9ee5cc5a6f8c0000)
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS reply setclientid_confirm: -10008
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS call setclientid auth=UNIX, 'Linux NFSv4.0 10.17.25.152/10.17.25.27 tcp'
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS reply setclientid: 0
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS call setclientid_confirm auth=UNIX, (client ID 9ee5cc5a708c0000)
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS reply setclientid_confirm: -10008
Aug 31 16:26:08 SEFRAPP05155 kernel: --> nfs_put_client({9})
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS: <-- nfs40_walk_client_list status = -10008
Aug 31 16:26:08 SEFRAPP05155 kernel: nfs4_get_rootfh: getroot error = 512
Aug 31 16:26:08 SEFRAPP05155 kernel: <-- nfs4_get_rootfh() = -512
Aug 31 16:26:08 SEFRAPP05155 kernel: --> nfs_free_server()
Aug 31 16:26:08 SEFRAPP05155 kernel: --> nfs_put_client({8})
Aug 31 16:26:08 SEFRAPP05155 kernel: <-- nfs_free_server()
Aug 31 16:26:08 SEFRAPP05155 kernel: <-- nfs4_create_server() = error -512
...
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS call setclientid_confirm auth=UNIX, (client ID 9ee5cc5a6f8c0000)
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS reply setclientid_confirm: -10008
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS call setclientid auth=UNIX, 'Linux NFSv4.0 10.17.25.152/10.17.25.27 tcp'
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS reply setclientid: 0
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS call setclientid_confirm auth=UNIX, (client ID 9ee5cc5a708c0000)
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS reply setclientid_confirm: -10008
Aug 31 16:26:08 SEFRAPP05155 kernel: --> nfs_put_client({9})
Aug 31 16:26:08 SEFRAPP05155 kernel: NFS: <-- nfs40_walk_client_list status = -10008
Aug 31 16:26:08 SEFRAPP05155 kernel: nfs4_get_rootfh: getroot error = 512
Aug 31 16:26:08 SEFRAPP05155 kernel: <-- nfs4_get_rootfh() = -512
Aug 31 16:26:08 SEFRAPP05155 kernel: --> nfs_free_server()
Aug 31 16:26:08 SEFRAPP05155 kernel: --> nfs_put_client({8})
Aug 31 16:26:08 SEFRAPP05155 kernel: <-- nfs_free_server()
Aug 31 16:26:08 SEFRAPP05155 kernel: <-- nfs4_create_server() = error -512
Aug 31 16:26:09 SEFRAPP05155 kernel: NFS call setclientid_confirm auth=UNIX, (client ID 9ee5cc5a708c0000)
Aug 31 16:26:09 SEFRAPP05155 kernel: NFS reply setclientid_confirm: -10008
Aug 31 16:26:09 SEFRAPP05155 kernel: NFS call setclientid auth=UNIX, 'Linux NFSv4.0 10.17.25.152/10.17.25.27 tcp'
Aug 31 16:26:09 SEFRAPP05155 kernel: NFS reply setclientid: 0
Aug 31 16:26:09 SEFRAPP05155 kernel: NFS call setclientid_confirm auth=UNIX, (client ID 9ee5cc5a718c0000)
Aug 31 16:26:09 SEFRAPP05155 kernel: NFS reply setclientid_confirm: -10008
Aug 31 16:26:09 SEFRAPP05155 kernel: --> nfs_put_client({8})
Aug 31 16:26:09 SEFRAPP05155 kernel: NFS: <-- nfs40_walk_client_list status = -10008
Eu vejo esses tipos de linhas de log repetidas.
Quando executo rpcinfo -p 10.17.25.27 no cliente, vejo:
program vers proto port service
100000 4 tcp 111 portmapper
100000 3 tcp 111 portmapper
100000 2 tcp 111 portmapper
100000 4 udp 111 portmapper
100000 3 udp 111 portmapper
100000 2 udp 111 portmapper
100024 1 udp 662 status
100005 1 udp 892 mountd
100024 1 tcp 662 status
100005 1 tcp 892 mountd
100005 2 udp 892 mountd
100005 2 tcp 892 mountd
100005 3 udp 892 mountd
100005 3 tcp 892 mountd
100003 3 tcp 2049 nfs
100003 4 tcp 2049 nfs
100227 3 tcp 2049 nfs_acl
100003 3 udp 2049 nfs
100003 4 udp 2049 nfs
100227 3 udp 2049 nfs_acl
100021 1 udp 32768 nlockmgr
100021 3 udp 32768 nlockmgr
100021 4 udp 32768 nlockmgr
100021 1 tcp 32768 nlockmgr
100021 3 tcp 32768 nlockmgr
100021 4 tcp 32768 nlockmgr
Observe que a montagem com a versão 3 do NFS funciona, mas eu preciso de recursos da versão 4, por isso não posso fazer o downgrade para essa versão mais antiga do protocolo.
Estou em CentOS 7.2 com o kernel 3.10.0-514.26.2.el7.x86_64
Não descobrimos por que nosso NFS caiu repentinamente e após a reinicialização, os clientes não conseguiram montar nenhum sistema de arquivos exportado. Tentamos reiniciar todos os servidores, para abrir todas as portas entre as máquinas, mas nada funcionou.
No entanto, conseguimos fazer com que funcionasse novamente depois que atualizamos o kernel da versão 3.10.0-514.26.2.el7.x86_64 para 3.10.0-862.11.6.el7.x86_64