把数据库删掉，在把数据库设备删掉，重建设备和库后，load出错

Rock_baggio

服务器：hpux+sybase12.5
把数据库删掉，在把数据库设备删掉，重启服务，重建设备和库后，load出错，日志如下：
Jul 20 10:05:28 2004: Backup Server: 4.149.1.2: Using Synchronous i/o for device '/res/res_nj/res_sz_data.dat'.
Jul 20 10:05:28 2004: Backup Server: 4.149.1.2: Using Synchronous i/o for device '/dbfs/res_sz_test/res_sz_log.dat'.
Jul 20 10:05:29 2004: Backup Server: 6.28.1.1: Dumpfile name '_accept04190081AE' section number 1 mounted on disk file '/res_upload/res/gis_check_accept2004_7_8.db'
Jul 20 10:05:29 2004: D00: SYBMULTBUF WARNING: Cannot use Asynchronous I/O for database device '/res/res_nj/res_sz_data.dat' because either the device type does not allow it or because the Adaptive Server Enterprise is using Synchronous I/O.  Backup Server is using Synchronous I/O for this device which will cause significant performance degradation.
Jul 20 10:05:29 2004: D00: SYBMULTBUF WARNING: Cannot use Asynchronous I/O for database device '/dbfs/res_sz_test/res_sz_log.dat' because either the device type does not allow it or because the Adaptive Server Enterprise is using Synchronous I/O.  Backup Server is using Synchronous I/O for this device which will cause significant performance degradation.
Jul 20 10:05:29 2004: Backup Server: 1.1.2.26: Memory allocation failed in bs_remap_runlist for a M_IO_LIST allocation.  This DUMP or LOAD session must exit.
Jul 20 10:05:29 2004: Backup Server: 6.32.2.3: /res_upload/res/gis_check_accept2004_7_8.db: volume not valid or not requested (server: , session id: 5.)
Jul 20 10:05:29 2004: Backup Server: 1.14.2.4: Unrecoverable I/O or volume error.  This DUMP or LOAD session must exit.

注：之前作load正常

Rock_baggio

补充：原来的数据库设备有6个，重建后的数据库设备只有两个，一个数据设备，一个日志设备，并且前后的设备大小没有改变

chenfeng825

这个错误和设备个数无关！
但同async IO有关，性能会很差的，调整os的文件系统参数
至于Load错误，是内存段分配异常，reboot backup server应该能够解决

Rock_baggio

试过了，reboot backup server后问题依旧，于是reboot server，结果提示：
Database <4>; cannot be opened - it is in the middle of a load.
server  Unable to proceed with the recovery of dbid <4>; because of previous errors.  Continuing with the next database.

dbid<4>;就是要load的库，服务启好后，库dbid<4>;用不了，只好update sysdatabases set  status=-32768 where dbid=4，再reboot server然后reboot backup server，最后load，问题依旧。日志中还是那个出错提示。

chenfeng825

dbcc dropdb掉有问题的db看看

Rock_baggio

问题好像不出在数据库上，请注意标题，这个库是我新建的，而且库出问题是因为load失败后，reboot server后所致。
另外，实在不行我想把服务卸载了重建，但不知如何操作，还望斑竹指点！
服务器：hpux 11.0+sybase ASE12.5

Rock_baggio

现在是把整个服务都重建了，也就是新建服务、设备、数据库，但load时问题依旧，日志：
Jul 21 15:45:43 2004: Backup Server: 4.149.1.2: Using Synchronous i/o for device '/res/res_nj/res_nj_data.dat'.
Jul 21 15:45:43 2004: Backup Server: 4.149.1.2: Using Synchronous i/o for device '/dbfs/res_sz_test/res_nj_log.dat'.
Jul 21 15:45:43 2004: Backup Server: 6.28.1.1: Dumpfile name '_accept04190081AE' section number 1 mounted on disk file '/res_upload/res/gis_check_accept2004_7_8.db'
Jul 21 15:45:43 2004: D00: SYBMULTBUF WARNING: Cannot use Asynchronous I/O for database device '/res/res_nj/res_nj_data.dat' because either the device type does not allow it or because the Adaptive Server Enterprise is using Synchronous I/O.  Backup Server is using Synchronous I/O for this device which will cause significant performance degradation.
Jul 21 15:45:43 2004: D00: SYBMULTBUF WARNING: Cannot use Asynchronous I/O for database device '/dbfs/res_sz_test/res_nj_log.dat' because either the device type does not allow it or because the Adaptive Server Enterprise is using Synchronous I/O.  Backup Server is using Synchronous I/O for this device which will cause significant performance degradation.
Jul 21 15:45:44 2004: Backup Server: 1.1.2.26: Memory allocation failed in bs_remap_runlist for a M_IO_LIST allocation.  This DUMP or LOAD session must exit.
Jul 21 15:45:44 2004: Backup Server: 6.32.2.3: /res_upload/res/gis_check_accept2004_7_8.db: volume not valid or not requested (server: , session id: 5.)
Jul 21 15:45:44 2004: Backup Server: 1.14.2.4: Unrecoverable I/O or volume error.  This DUMP or LOAD session must exit.
然后数据库就挂起了，看日志好像是说内存分配的问题，但就是不知道怎么解决，急啊。

joey

原来有6个设备，现在为何只用2个设备？？？
如果现在用6个设备呢？load 有问题吗？

laoma1994

because either the device type does not allow it or because the Adaptive Server Enterprise is using Synchronous I/O