Problems with On tape

This is a multi-part message in MIME format. ------=_NextPart_000_0029_01BE90B4.240D0A60 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable We have the On-Line 7.30 and we have executed "ontape -s" at level 0. When we execute "ontape -r", dbspaces aren't recovered and the On-Line = don't initializes. Can everyone help us? ------=_NextPart_000_0029_01BE90B4.240D0A60 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN"> <HTML> <HEAD> <META content=3Dtext/html;charset=3Diso-8859-1 = http-equiv=3DContent-Type> <META content=3D'"MSHTML 4.72.2106.6"' name=3DGENERATOR> </HEAD> <BODY bgColor=3D#ffffff> <DIV><FONT color=3D#000000 size=3D2>We have the On-Line 7.30 and we have = executed=20 &quot;ontape -s&quot; at level 0.</FONT></DIV> <DIV><FONT color=3D#000000 size=3D2>When we execute &quot;ontape = -r&quot;, dbspaces=20 aren't recovered and the On-Line don't initializes.</FONT></DIV> <DIV><FONT color=3D#000000 size=3D2></FONT>&nbsp;</DIV> <DIV><FONT color=3D#000000 size=3D2>Can everyone help = us?</FONT></DIV></BODY></HTML> ------=_NextPart_000_0029_01BE90B4.240D0A60--

Federica Pavese wrote: > > We have the On-Line 7.30 and we have executed "ontape -s" at level 0. > When we execute "ontape -r", dbspaces aren't recovered and the On-Line = > don't initializes. > > Can everyone help us? You do not post your platform, it would help. On our platform, DGUX, it takes a long time, relative to say Solaris, to connect to shared memory when a process is started. Since we normally run over 160 VPs (28 CPU, 20 tli, 120 aio) and have a large shared memory image (900MB+) it normally takes over 15 minutes to bring the instances online. When doing a restore the ontape gets tired of waiting for the engine to reach 'recovery' mode and times out exiting without restoring. To solve this problem I reduce the number of CPU VPs to 4, the number of AIO VPs to 10 and reduce the number of buffers from 200,000 to 2,000 before attempting a restore. This allows the engine to start up before ontape gives up on it. Then once the restore is complete be diligent to bring the engine to online mode before shutting the limited instance down and restoring the ONCONFIG parameters for a production startup otherwise the engine will not start and you will have to begin the entire restore over again. Art S. Kagel