and 这两个看起来都像pg_user视图已损坏。 似乎很难相信同样的腐败可能会同时影响两个数据库。您是否在所有数据库中看到同样的问题(如果您还有模板1和电影院)?如果是这样的话,我想知道postgres可执行文件是否已损坏---它对于系统无法破坏未被修改的文件非常常见,但是也许它可能发生。如果是这样,你可以通过快速重新安装RPM来解决它。 问候,tom lane These both look like the pg_user view is corrupted. It seems hard to believe that identical corruption could have affected two databases at once. Do you see this same problem with all your databases (if you have any more than template1 and cinema)? If so, I wonder whether the postgres executable got corrupted --- it''d not be very common for a system failure to corrupt files that aren''t being modified, but perhaps it could happen. If it is that, you could fix it with a quick reinstall of the RPM. regards, tom lane - --------------------------(广播结束)------------------- -------- 提示6:您是否搜索了我们的列表档案? http://archives.postgresql.org---------------------------(end of broadcast)---------------------------TIP 6: Have you searched our list archives? http://archives.postgresql.org 嗯,这很奇怪。我们能看到从pg_rewrite中选择*的结果吗?其中ev_class =''pg_user'':: regclass; 和 选择pg_get_viewdef(''pg_user'':: regclass); 你可能需要做一个回滚。启动后让psql退出其b / b 事务中止状态,但是我希望 中的第一个能够正常运行。第二个可能会失败,但是我们应该确认......确认... ... 问候,汤姆小巷 ---------------------------(广播结束)------------------ --------- 提示1:订阅和取消订阅命令转到 ma ** *****@postgresql.orgHm, that''s very odd. Could we see the results ofselect * from pg_rewrite where ev_class = ''pg_user''::regclass;andselect pg_get_viewdef(''pg_user''::regclass);You''ll probably need to do a "rollback" to get psql out of itstransaction-aborted state after startup, but then I''d expect thefirst of these to work. The second will likely fail, but we shouldconfirm that ...regards, tom lane---------------------------(end of broadcast)---------------------------TIP 1: subscribe and unsubscribe commands go to ma*******@postgresql.org 这篇关于错误:nodeRead:找不到'}'[x2]的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持! 上岸,阿里云!
09-06 01:34