[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] RE: Re: [Xen-users] Live migration problem
One thing I don't understand is when I look at xc_linux_save.c's suspend_and_state function - it appears it does: xcio_suspend_domain(ioctxt); retry: ... stuff tries to see if xcio_suspend_domain worked - is that correct? Should the xcio_suspend_domain() call be after the retry: label? Or does that xcio_suspend_domain call guarantee the message is delivered and the code under retry: is just waiting for the state to now change? Second, I see it tries 100 times with a 10,000microsecond sleep inbetween. So it only waits for 1 second for the domain to suspend. I realize 1 second is a really long time in terms of computing. But I'm wondering what all conditions must be true for it to be able to suspend a domain. Could there legitimately be times when it would take longer than a second for suspension? -- Ray > -----Original Message----- > From: Cole, Ray > Sent: Wednesday, August 24, 2005 5:33 PM > To: 'xen-users@xxxxxxxxxxxxxxxxxxx' > Subject: RE: Re: [Xen-users] Live migration problem > > This is a snippet from the originating host's xfrd.log when the failure > happened. Meanwhile the other side's xfrd.log looked good until it got an > 'Error when reading from state file'. > > > -- Ray > > > ----- > > > 1: sent 45431, skipped 3721, > > 1: sent 45431, skipped 3721, delta 16818ms, dom0 23%, target 9%, sent > 88Mb/s, dirtied 8Mb/s 4557 pages > [1124922400.612070] Saving memory pages: iter 2 0% > Saving memory pages: iter 2 0% 22% > 22% 47% > 47% 71% > 71% > 2: sent 4054, skipped 500, > > 2: sent 4054, skipped 500, delta 1571ms, dom0 22%, target 21%, sent 84Mb/s, > dirtied 64Mb/s 3116 pages > [1124922402.183740] Saving memory pages: iter 3 0% > Saving memory pages: iter 3 0% 31% > 31% 73% > 73% > 3: sent 2693, skipped 420, > > 3: sent 2693, skipped 420, delta 1022ms, dom0 23%, target 19%, sent 86Mb/s, > dirtied 65Mb/s 2053 pages > [1124922403.205810] Saving memory pages: iter 4 0% > Saving memory pages: iter 4 0% 48% > 48% > 4: sent 1874, skipped 176, > > 4: sent 1874, skipped 176, delta 705ms, dom0 23%, target 17%, sent 87Mb/s, > dirtied 61Mb/s 1315 pages > [1124922403.911777] Saving memory pages: iter 5 0% > Saving memory pages: iter 5 0% 87% > 87% > 5: sent 1107, skipped 201, > > 5: sent 1107, skipped 201, delta 470ms, dom0 21%, target 40%, sent 77Mb/s, > dirtied 128Mb/s 1846 pages > [1124922404.382522] Saving memory pages: iter 6 0% > Saving memory pages: iter 6 0% 63% > 63% > 6: sent 1491, skipped 349, > > 6: sent 1491, skipped 349, delta 609ms, dom0 22%, target 46%, sent 80Mb/s, > dirtied 142Mb/s 2647 pages > [1124922404.992153] Saving memory pages: iter 7 0% > Saving memory pages: iter 7 0% 38% > 38% 79% > 79% > 7: sent 2348, skipped 295, > > 7: sent 2348, skipped 295, delta 890ms, dom0 23%, target 18%, sent 86Mb/s, > dirtied 102Mb/s 2797 pages > [1124922405.882768] Saving memory pages: iter 8 0% > Saving memory pages: iter 8 0% 38% > 38% 84% > 84% > 8: sent 2409, skipped 384, > > 8: sent 2409, skipped 384, delta 860ms, dom0 24%, target 6%, sent 91Mb/s, > dirtied 27Mb/s 713 pages > [1124922406.742903] Saving memory pages: iter 9 0% > Saving memory pages: iter 9 0% > 9: sent 624, skipped 83, > > 9: sent 624, skipped 83, delta 230ms, dom0 23%, target 9%, sent 88Mb/s, > dirtied 58Mb/s 410 pages > [1124922406.973505] Saving memory pages: iter 10 0% > Saving memory pages: iter 10 0% > 10: sent 404, skipped 0, > > 10: sent 404, skipped 0, delta 142ms, dom0 26%, target 6%, sent 93Mb/s, > dirtied 51Mb/s 223 pages > [1124922407.118014] Saving memory pages: iter 11 0% > Saving memory pages: iter 11 0% > 11: sent 127, skipped 89, > > 11: sent 127, skipped 89, delta 47ms, dom0 29%, target 6%, sent 88Mb/s, > dirtied 150Mb/s 216 pages > [1124922407.163792] Saving memory pages: iter 12 0%> > Saving memory pages: iter 12 0% > 12: sent 210, skipped 0, > > 12: sent 210, skipped 0, delta 78ms, dom0 25%, target 10%, sent 88Mb/s, > dirtied 132Mb/s 315 pages > [1124922407.242383] Saving memory pages: iter 13 0% > Saving memory pages: iter 13 0% > 13: sent 309, skipped 0, > > 13: sent 309, skipped 0, delta 113ms, dom0 25%, target 7%, sent 89Mb/s, > dirtied 91Mb/s 317 pages > [1124922407.355431] Saving memory pages: iter 14 0% > Saving memory pages: iter 14 0% > 14: sent 310, skipped 0, > > 14: sent 310, skipped 0, delta 113ms, dom0 25%, target 7%, sent 89Mb/s, > dirtied 82Mb/s 283 pages > [1124922407.468703] Saving memory pages: iter 15 0% > Saving memory pages: iter 15 0% > 15: sent 277, skipped 0, > > 15: sent 277, skipped 0, delta 99ms, dom0 26%, target 7%, sent 91Mb/s, > dirtied 94Mb/s 287 pages > [1124922407.568408] Saving memory pages: iter 16 0% > Saving memory pages: iter 16 0% > 16: sent 281, skipped 0, > > 16: sent 281, skipped 0, delta 102ms, dom0 26%, target 8%, sent 90Mb/s, > dirtied 107Mb/s 334 pages > [1124922407.671120] Saving memory pages: iter 17 0% > Saving memory pages: iter 17 0% > 17: sent 243, skipped 86, > > 17: sent 243, skipped 86, delta 93ms, dom0 25%, target 13%, sent 85Mb/s, > dirtied 135Mb/s 385 pages > [1124922407.764443] Saving memory pages: iter 18 0% > Saving memory pages: iter 18 0% > 18: sent 378, skipped 0, > > 18: sent 378, skipped 0, delta 144ms, dom0 24%, target 11%, sent 86Mb/s, > dirtied 82Mb/s 363 pages > [1124922407.908636] Saving memory pages: iter 19 0% > Saving memory pages: iter 19 0% > 19: sent 355, skipped 0, > > 19: sent 355, skipped 0, delta 130ms, dom0 26%, target 6%, sent 89Mb/s, > dirtied 71Mb/s 283 pages > [1124922408.038797] Saving memory pages: iter 20 0% > Saving memory pages: iter 20 0% > 20: sent 185, skipped 92, > > 20: sent 185, skipped 92, delta 106ms, dom0 17%, target 81%, sent 57Mb/s, > dirtied 218Mb/s 707 pages > [1124922408.145378] Saving memory pages: iter 21 0% > Saving memory pages: iter 21 0% > 21: sent 619, skipped 83, > > 21: sent 619, skipped 83, delta 229ms, dom0 23%, target 13%, sent 88Mb/s, > dirtied 94Mb/s 657 pages > [1124922408.375295] Saving memory pages: iter 22 0% > Saving memory pages: iter 22 0% > 22: sent 651, skipped 0, > > 22: sent 651, skipped 0, delta 231ms, dom0 25%, target 2%, sent 92Mb/s, > dirtied 18Mb/s 130 pages > [1124922408.606305] Saving memory pages: iter 23 0% > Saving memory pages: iter 23 0% > 23: sent 124, skipped 0, > > 23: sent 124, skipped 0, delta 45ms, dom0 31%, target 6%, sent 90Mb/s, > dirtied 120Mb/s 166 pages > [1124922408.651758] Saving memory pages: iter 24 0% > Saving memory pages: iter 24 0% > 24: sent 159, skipped 0, > > 24: sent 159, skipped 0, delta 57ms, dom0 28%, target 8%, sent 91Mb/s, > dirtied 143Mb/s 249 pages > [1124922408.709624] Saving memory pages: iter 25 0% > Saving memory pages: iter 25 0% > 25: sent 243, skipped 0, > > 25: sent 243, skipped 0, delta 102ms, dom0 21%, target 79%, sent 78Mb/s, > dirtied 57Mb/s 178 pages > [1124922408.812289] Saving memory pages: iter 26 0% > Saving memory pages: iter 26 0% > 26: sent 168, skipped 4, > > 26: sent 168, skipped 4, delta 70ms, dom0 21%, target 78%, sent 78Mb/s, > dirtied 20Mb/s 43 pages > [1124922408.883198] Saving memory pages: iter 27 0% > Saving memory pages: iter 27 0% > 27: sent 39, skipped 4, > > 27: sent 39, skipped 4, [DEBUG] Conn_sxpr> > (xfr.err 22)[DEBUG] Conn_sxpr< err=0 > Retry suspend domain (0) > ...repeated many times... > Retry suspend domain (0) > Retry suspend domain (0) > Unable to suspend domain. (0) > Unable to suspend domain. (0) > Domain appears not to have suspended: 0 > Domain appears not to have suspended: 0 > 6638 [WRN] XFRD> Transfer errors: > 6638 [WRN] XFRD> state=XFR_STATE err=1 > 6638 [INF] XFRD> Xfr service err=1 > > -- Ray > > -----Original Message-----> > From: Cole, Ray > Sent: Wednesday, August 24, 2005 11:32 AM > To: 'xen-users@xxxxxxxxxxxxxxxxxxx' > Subject: Re: [Xen-users] Live migration problem > > I'm having the exact same issue here with 2.0.7. I'm using RedHat AS 4 > on one machine, Fedora Core 4 on the other using xen 2.0.7 built from source. > The guest OS is RH AS 4. I'm also able to migrate a number of times in a > row without a problem, but periodically get the exact same errors that Steven > Yelton posted. If my guest OS is more or less idle it seems to be less > likely to happen. But if I start a big compile job and trying to do the > migrate it is more likely to fail. > > -- Ray _______________________________________________ Xen-users mailing list Xen-users@xxxxxxxxxxxxxxxxxxx http://lists.xensource.com/xen-users
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |