O upgrade do IDPA vCenter apresenta falha com o erro "vCSACliInstallLogger - ERROR - vCenter Server Data Export failed"

摘要: Este artigo da base de conhecimento fornece orientações sobre a falha do vCenter Upgrade causada devido à falha na exportação do banco de dados do vCenter. Isto pode ser devido à corrupção do banco de dados do vCenter. ...

本文适用于 本文不适用于 本文并非针对某种特定的产品。 本文并非包含所有产品版本。
查看其他资源

症状



No ACM, você pode notar que o vCenter mostra como apresentou falha.

Analisando o registro do vCenter Upgrade no diretório "/data01/tmp/patch/logs/VCSA", os seguintes erros podem ser vistos no registro do vcsaclientinstaller:

Progress: 42% Exporting VMware vCenter Server Database data...
2020-03-26 03:41:00,145 - vCSACliInstallLogger - ERROR - vCenter Server Data Export failed.
2020-03-26 03:41:00,145 - vCSACliInstallLogger - ERROR - Upgrade failed:
Status: ERROR
Progress: 42% Exporting VMware vCenter Server Database data...
Error:
Problem Id: None
Component key: upgrade_framework
Detail:
Error while exporting source SET tables: pg_dump: Error message from server: ERRO:  invalid page in block 9 of relation base/16384/25164

Resolution: Read the log files and address the issues found in your source SET tables.
2020-03-26 03:41:00,145 - vCSACliInstallLogger - INFO - See /data01/tmp/patch/logs/vclogs/monitor-firstboot-progress-oczvL_.json file on local machine or /var/log/vmware/upgrade/export.json file on remote machine for details.


Além disso, no servidor do vCenter, os registros de upgrade em "/var/log/vmware/upgrade" também podem ser verificados e podem exibir a seguinte mensagem de erro:

VMware License Service data...
2020-03-26 17:09:01,732 - vCSACliInstallLogger - INFO - vCenter Server Data Export: Progress: 42% Exporting VMware vCenter Server Database data...
2020-03-26 17:09:12,296 - vCSACliInstallLogger - ERROR - vCenter Server Data Export failed.
2020-03-26 17:09:12,296 - vCSACliInstallLogger - ERROR - Upgrade failed:
Status: ERROR
Progress: 42% Exporting VMware vCenter Server Database data...
Error:
Problem Id: None
Component key: upgrade_framework
Detail:
Error while exporting source SET tables: pg_dump: Error message from server: ERRO:  invalid page in block 9 of relation base/16384/25164

Resolution: Read the log files and address the issues found in your source SET tables.
2020-03-26 17:09:12,296 - vCSACliInstallLogger - INFO - See /data01/tmp/patch/logs/vclogs/monitor-firstboot-progress-tmcn2I.json file on local machine or /var/log/vmware/upgrade/export.json file on remote machine for details.
2020-03-26 17:09:12,305 - vCSACliInstallLogger - INFO - Gathering VC support log bundle. This can take a few minutes.
2020-03-26 17:09:12,583 - vCSACliInstallLogger - INFO - Uninstalling vmware-upgrade-requirements-* on 192.168.100.99.
2020-03-26 17:09:13,881 - vCSACliInstallLogger - ERROR - (u'Failed to uninstall %(rpm)s on %(host)s: %(error)s', {'host': u'192.168.100.99', 'rpm': u'vmware-upgrade-requirements-*', 'error': CommandError(u"Failed to run and wait for command in guest with error 'Command '/bin/rpm' exited with non-zero status 1'",)})


No vCenter, o arquivo postgres.log mostra a seguinte mensagem de erro: 

816 UTC 5e7ce0cc.7647 24294965  LOG: automatic analyze of table "VCDB.vc.vpx_event_arg" system usage: CPU 0.02s/1.53u sec elapsed 6.15 sec
2020-03-26 17:08:19.502 UTC 5e7ce180.2530 24295002  LOG: automatic analyze of table "VCDB.vc.vpx_event" system usage: CPU 0.00s/2.40u sec elapsed 2.79 sec
2020-03-26 17:08:58.326 UTC 5e7c377f.3892 0 VCDB vc LOG: unexpected EOF on client connection with an open transaction
2020-03-26 17:08:58.326 UTC 5e7c3796.3a7b 0 VCDB vc LOG: unexpected EOF on client connection with an open transaction
2020-03-26 17:08:58.327 UTC 5e7c3796.3a81 0 VCDB vc LOG: unexpected EOF on client connection with an open transaction
2020-03-26 17:09:09.530 UTC 5e7ce1b3.3c67 0 VCDB postgres ERROR: invalid page in block 9 of relation base/16384/25164
2020-03-26 17:09:09.530 UTC 5e7ce1b3.3c67 0 VCDB postgres STATEMENT: COPY vc.vpx_topn_past_month (counter_id, sample_time, integral_value) TO stdout;
SET zero_damaged_pages = on; select * from <corrupted table>; SET zero_damaged_pages = off;

VCDB=# select counter_id from vpx_topn_past_month;
ERROR: invalid page in block 9 of relation base/16384/25164

原因

A corrupção do banco de dados Postgres do vCenter causou a falha do vCenter Upgrade durante a tarefa de exportação do banco de dados. 

解决方案

Etapa 1: Faça login na IU do IDPA vCenter.

Etapa 2: Crie um clone da VM do vCenter.
  • Clique com o botão direito em VCSA VM --> Clone to Virtual Machine. 
  • Escolha o nome do clone e outros detalhes e clique em OK. 

Etapa 3: Execute uma reversão para a VM do vCenter na IU do ACM. 
  • Clique na caixa de seleção Troubleshooting Options. Clique em Rollback for vCenter Upgrade.  

Etapa 4: SSH para vCenter VM e interromper o serviço vpxd.
  • service-control --stop vpxd

Etapa 5: Faça login no banco de dados Postgres do vCenter.
  • /opt/vmware/vpostgres/current/bin/psql -d VCDB -U postgres
Etapa 6: Execute as seguintes consultas/comandos no banco de dados Postgres do vCenter:
  • SET zero_damaged_pages = on;
  • VACUUM FULL ANALYSE VERBOSE;
  • REINDEX table vpx_topn_past_month;
  • select * from vpx_topn_past_month;
  • SET zero_damaged_pages = off;
Etapa 7: Vá para a interface do usuário do ACM, clique em Retry for vCenter Upgrade e monitore. 


    其他信息

    Envolva o suporte do VMware se o problema não for resolvido depois de seguir as etapas acima. 

    受影响的产品

    Integrated Data Protection Appliance Family

    产品

    PowerProtect DP4400, PowerProtect DP5300, PowerProtect DP5800, PowerProtect DP8300, PowerProtect DP8800, PowerProtect Data Protection Software, Integrated Data Protection Appliance Family, PowerProtect Data Protection Hardware , Integrated Data Protection Appliance Software ...
    文章属性
    文章编号: 000173289
    文章类型: Solution
    上次修改时间: 27 5月 2025
    版本:  3
    从其他戴尔用户那里查找问题的答案
    支持服务
    检查您的设备是否在支持服务涵盖的范围内。