Hi Maxim,
Thanks for your input.
I want to load lost data into cache in case of lost partitions happened. I
am handling this cache loading in event listener(which is listening
to *EVT_CACHE_REBALANCE_PART_DATA_LOST
*event).
I am taking help of IgniteCache.lostPartitions() method to get the lost
partitions. I am going to pass this partition list to cache loader.  Issue
is that I am receiving this event call per partition. But I want to start
cache loading for all lost partitions .

After loading data successfully  next step is to reset lost partition list
using resetLostPartitions () method.

Can you suggest any better way to reload lost data due to lost partitions .

Thanks,
Akash

On Thu, Sep 19, 2019 at 12:48 AM Maxim Muzafarov <[email protected]> wrote:

> Hello,
>
> 1) I'm not sure that I've cached your idea right, but is this method
> are you looking for?
> org.apache.ignite.IgniteCache#lostPartitions [1]
>
> 2) I doubt that it is possible since resetLostPartitions method
> accepts only cache names.
>
> [1]
> https://ignite.apache.org/releases/latest/javadoc/org/apache/ignite/IgniteCache.html#lostPartitions--
>
> On Tue, 17 Sep 2019 at 16:14, Akash Shinde <[email protected]> wrote:
> >
> > Can someone please help me on this?
> >
> > Thanks,
> > Akash
> >
> > On Tue, Sep 17, 2019 at 10:40 AM Akash Shinde <[email protected]>
> wrote:
> >>
> >> Hi,
> >> I am trying to recover lost data in case of partition loss.
> >> In my ignite configuration native persistence is off.
> >> I have started event listener on EVT_CACHE_REBALANCE_PART_DATA_LOST
> event. This listener will get lost partition list using
> cache.lostPartitions() method.
> >> The issue is that listener gets call per partition. So if there 100
> partition loss due to single node termination then 100 time this listener
> will get called and last multiple calls to the listener will fetch all lost
> partition list.
> >>
> >> Lets take a scenario:
> >> Started two server nodes  Node A and Node B.  Started cache with
> partition mode and the number of backup set to 0 in order to facilitate
> simulation of partition loss scenarios
> >> Started event listener on both node listening  to
> EVT_CACHE_REBALANCE_PART_DATA_LOST  event.
> >>
> >> Number of partitions on node A = 500
> >> Number of partitions on node B = 524
> >>
> >> Now stop node B. After termination of node B listener running on node A
> gets call multiple time per partition.
> >> I have printed logs on listener
> >>
> >> primary partition size after loss:1024
> >> Lost partion Nos.1
> >> IgniteThread [compositeRwLockIdx=1, stripe=-1, plc=-1,
> name=exchange-worker-#42%springDataNode%]::[0]
> >> Event Detail:CacheRebalancingEvent [cacheName=ASSET_GROUP_CACHE,
> part=0, discoNode=TcpDiscoveryNode
> [id=1bb17828-3556-499f-a4e6-98cfdc1d11fb, addrs=[0:0:0:0:0:0:0:1,
> 10.113.14.98, 127.0.0.1], sockAddrs=[], discPort=47501, order=2,
> intOrder=2, lastExchangeTime=1568357181089, loc=false,
> ver=2.6.0#20180710-sha1:669feacc, isClient=false], discoEvtType=12,
> discoTs=1568357376683, discoEvtName=NODE_FAILED, nodeId8=499400ac,
> msg=Cache rebalancing event., type=CACHE_REBALANCE_PART_DATA_LOST,
> tstamp=1568357376714]
> >> primary partition size after loss:1024
> >> Lost partion Nos.2
> >> IgniteThread [compositeRwLockIdx=1, stripe=-1, plc=-1,
> name=exchange-worker-#42%springDataNode%]::[0, 1]
> >> Event Detail:CacheRebalancingEvent [cacheName=ASSET_GROUP_CACHE,
> part=1, discoNode=TcpDiscoveryNode
> [id=1bb17828-3556-499f-a4e6-98cfdc1d11fb, addrs=[0:0:0:0:0:0:0:1,
> 10.113.14.98, 127.0.0.1], sockAddrs=[], discPort=47501, order=2,
> intOrder=2, lastExchangeTime=1568357181089, loc=false,
> ver=2.6.0#20180710-sha1:669feacc, isClient=false], discoEvtType=12,
> discoTs=1568357376683, discoEvtName=NODE_FAILED, nodeId8=499400ac,
> msg=Cache rebalancing event., type=CACHE_REBALANCE_PART_DATA_LOST,
> tstamp=1568357376726]
> >> primary partition size after loss:1024
> >> Lost partion Nos.3
> >> IgniteThread [compositeRwLockIdx=1, stripe=-1, plc=-1,
> name=exchange-worker-#42%springDataNode%]::[0, 1, 2]
> >> Event Detail:CacheRebalancingEvent [cacheName=ASSET_GROUP_CACHE,
> part=2, discoNode=TcpDiscoveryNode
> [id=1bb17828-3556-499f-a4e6-98cfdc1d11fb, addrs=[0:0:0:0:0:0:0:1,
> 10.113.14.98, 127.0.0.1], sockAddrs=[], discPort=47501, order=2,
> intOrder=2, lastExchangeTime=1568357181089, loc=false,
> ver=2.6.0#20180710-sha1:669feacc, isClient=false], discoEvtType=12,
> discoTs=1568357376683, discoEvtName=NODE_FAILED, nodeId8=499400ac,
> msg=Cache rebalancing event., type=CACHE_REBALANCE_PART_DATA_LOST,
> tstamp=1568357376726]
> >> primary partition size after loss:1024
> >> Lost partion Nos.4
> >> IgniteThread [compositeRwLockIdx=1, stripe=-1, plc=-1,
> name=exchange-worker-#42%springDataNode%]::[0, 1, 2, 4]
> >> Event Detail:CacheRebalancingEvent [cacheName=ASSET_GROUP_CACHE,
> part=4, discoNode=TcpDiscoveryNode
> [id=1bb17828-3556-499f-a4e6-98cfdc1d11fb, addrs=[0:0:0:0:0:0:0:1,
> 10.113.14.98, 127.0.0.1], sockAddrs=[], discPort=47501, order=2,
> intOrder=2, lastExchangeTime=1568357181089, loc=false,
> ver=2.6.0#20180710-sha1:669feacc, isClient=false], discoEvtType=12,
> discoTs=1568357376683, discoEvtName=NODE_FAILED, nodeId8=499400ac,
> msg=Cache rebalancing event., type=CACHE_REBALANCE_PART_DATA_LOST,
> tstamp=1568357376736]
> >> primary partition size after loss:1024
> >> Lost partion Nos.5
> >> .
> >> .
> >> .
> >> .
> >> IgniteThread [compositeRwLockIdx=1, stripe=-1, plc=-1,
> name=exchange-worker-#42%springDataNode%]::[0, 1, 2, 4, 5, 6, 7, 11, 13,
> 17, 22, 26, 28, 29, 30, 33, 34, 37, 38, 41, 43, 45, 47, 48, 49, 50, 55, 58,
> 61, 62, 64, 65, 68, 70, 71, 75, 77, 79, 81, 82, 85, 87, 88, 89, 90, 93,
> 100, 101, 102, 104, 110, 112, 114, 116, 121, 123, 125, 126, 132, 133, 135,
> 137, 138, 139, 140, 144, 145, 146, 147, 149, 150, 151, 154, 156, 157, 158,
> 163, 164, 165, 169, 170, 172, 173, 176, 178, 180, 182, 183, 184, 185, 195,
> 196, 198, 199, 203, 204, 212, 213, 215, 217, 219, 220, 222, 223, 224, 226,
> 227, 230, 233, 234, 236, 237, 240, 242, 245, 248, 250, 251, 253, 255, 257,
> 258, 263, 265, 266, 267, 269, 270, 272, 273, 275, 276, 277, 278, 281, 282,
> 283, 287, 288, 292, 293, 295, 296, 297, 298, 300, 301, 302, 305, 308, 309,
> 310, 311, 313, 314, 315, 318, 319, 320, 322, 323, 324, 326, 327, 328, 329,
> 330, 331, 332, 333, 336, 340, 342, 344, 347, 348, 349, 351, 352, 353, 354,
> 355, 357, 362, 364, 369, 370, 371, 373, 374, 375, 376, 380, 382, 383, 387,
> 389, 394, 395, 396, 397, 398, 401, 402, 403, 407, 408, 409, 410, 411, 412,
> 413, 416, 417, 421, 424, 425, 427, 430, 431, 433, 435, 437, 438, 439, 440,
> 441, 442, 443, 445, 446, 452, 454, 455, 456, 459, 461, 463, 466, 470, 472,
> 474, 475, 476, 480, 481, 482, 484, 485, 489, 492, 494, 495, 496, 497, 498,
> 499, 501, 502, 503, 504, 505, 508, 510, 511, 512, 513, 514, 515, 516, 519,
> 523, 525, 526, 527, 529, 530, 531, 532, 535, 536, 539, 540, 541, 543, 545,
> 546, 550, 552, 553, 555, 557, 560, 569, 572, 573, 575, 576, 579, 582, 589,
> 591, 593, 594, 597, 599, 602, 603, 604, 605, 607, 608, 610, 612, 613, 614,
> 615, 616, 617, 619, 622, 624, 625, 626, 627, 630, 631, 632, 633, 634, 635,
> 636, 637, 638, 639, 640, 641, 642, 643, 645, 646, 647, 648, 649, 652, 653,
> 654, 656, 657, 660, 662, 663, 666, 668, 669, 670, 671, 679, 681, 683, 686,
> 688, 691, 693, 698, 701, 702, 703, 705, 706, 709, 712, 713, 716, 717, 719,
> 721, 723, 726, 730, 737, 738, 740, 741, 742, 745, 747, 750, 752, 755, 756,
> 759, 760, 761, 763, 764, 765, 766, 767, 768, 770, 771, 772, 777, 779, 785,
> 786, 789, 790, 792, 793, 794, 799, 801, 804, 811, 816, 818, 822, 823, 824,
> 825, 826, 827, 832, 833, 836, 838, 840, 841, 843, 844, 846, 850, 851, 852,
> 853, 855, 856, 858, 862, 864, 867, 872, 873, 876, 877, 878, 879, 883, 884,
> 886, 887, 890, 892, 895, 897, 898, 899, 900, 902, 903, 904, 905, 906, 907,
> 908, 910, 914, 916, 918, 919, 920, 921, 922, 925, 926, 928, 929, 933, 935,
> 936, 939, 940, 943, 945, 950, 951, 952, 953, 960, 961, 963, 964, 966, 967,
> 972, 973, 975, 977, 979, 980, 982, 983, 984, 985, 987, 989, 991, 992, 995,
> 996, 999, 1002, 1003, 1005, 1007, 1011, 1014, 1015, 1016, 1018, 1020, 1021]
> >> Event Detail:CacheRebalancingEvent [cacheName=ASSET_GROUP_CACHE,
> part=412, discoNode=TcpDiscoveryNode
> [id=1bb17828-3556-499f-a4e6-98cfdc1d11fb, addrs=[0:0:0:0:0:0:0:1,
> 10.113.14.98, 127.0.0.1], sockAddrs=[], discPort=47501, order=2,
> intOrder=2, lastExchangeTime=1568357181089, loc=false,
> ver=2.6.0#20180710-sha1:669feacc, isClient=false], discoEvtType=12,
> discoTs=1568357376683, discoEvtName=NODE_FAILED, nodeId8=499400ac,
> msg=Cache rebalancing event., type=CACHE_REBALANCE_PART_DATA_LOST,
> tstamp=1568357423500]
> >> primary partition size after loss:1024
> >> Lost partion Nos.524
> >> IgniteThread [compositeRwLockIdx=1, stripe=-1, plc=-1,
> name=exchange-worker-#42%springDataNode%]::[0, 1, 2, 4, 5, 6, 7, 11, 13,
> 17, 22, 26, 28, 29, 30, 33, 34, 37, 38, 41, 43, 45, 47, 48, 49, 50, 55, 58,
> 61, 62, 64, 65, 68, 70, 71, 75, 77, 79, 81, 82, 85, 87, 88, 89, 90, 93,
> 100, 101, 102, 104, 110, 112, 114, 116, 121, 123, 125, 126, 132, 133, 135,
> 137, 138, 139, 140, 144, 145, 146, 147, 149, 150, 151, 154, 156, 157, 158,
> 163, 164, 165, 169, 170, 172, 173, 176, 178, 180, 182, 183, 184, 185, 195,
> 196, 198, 199, 203, 204, 212, 213, 215, 217, 219, 220, 222, 223, 224, 226,
> 227, 230, 233, 234, 236, 237, 240, 242, 245, 248, 250, 251, 253, 255, 257,
> 258, 263, 265, 266, 267, 269, 270, 272, 273, 275, 276, 277, 278, 281, 282,
> 283, 287, 288, 292, 293, 295, 296, 297, 298, 300, 301, 302, 305, 308, 309,
> 310, 311, 313, 314, 315, 318, 319, 320, 322, 323, 324, 326, 327, 328, 329,
> 330, 331, 332, 333, 336, 340, 342, 344, 347, 348, 349, 351, 352, 353, 354,
> 355, 357, 362, 364, 369, 370, 371, 373, 374, 375, 376, 380, 382, 383, 387,
> 389, 394, 395, 396, 397, 398, 401, 402, 403, 407, 408, 409, 410, 411, 412,
> 413, 416, 417, 421, 424, 425, 427, 430, 431, 433, 435, 437, 438, 439, 440,
> 441, 442, 443, 445, 446, 452, 454, 455, 456, 459, 461, 463, 466, 470, 472,
> 474, 475, 476, 480, 481, 482, 484, 485, 489, 492, 494, 495, 496, 497, 498,
> 499, 501, 502, 503, 504, 505, 508, 510, 511, 512, 513, 514, 515, 516, 519,
> 523, 525, 526, 527, 529, 530, 531, 532, 535, 536, 539, 540, 541, 543, 545,
> 546, 550, 552, 553, 555, 557, 560, 569, 572, 573, 575, 576, 579, 582, 589,
> 591, 593, 594, 597, 599, 602, 603, 604, 605, 607, 608, 610, 612, 613, 614,
> 615, 616, 617, 619, 622, 624, 625, 626, 627, 630, 631, 632, 633, 634, 635,
> 636, 637, 638, 639, 640, 641, 642, 643, 645, 646, 647, 648, 649, 652, 653,
> 654, 656, 657, 660, 662, 663, 666, 668, 669, 670, 671, 679, 681, 683, 686,
> 688, 691, 693, 698, 701, 702, 703, 705, 706, 709, 712, 713, 716, 717, 719,
> 721, 723, 726, 730, 737, 738, 740, 741, 742, 745, 747, 750, 752, 755, 756,
> 759, 760, 761, 763, 764, 765, 766, 767, 768, 770, 771, 772, 777, 779, 785,
> 786, 789, 790, 792, 793, 794, 799, 801, 804, 811, 816, 818, 822, 823, 824,
> 825, 826, 827, 832, 833, 836, 838, 840, 841, 843, 844, 846, 850, 851, 852,
> 853, 855, 856, 858, 862, 864, 867, 872, 873, 876, 877, 878, 879, 883, 884,
> 886, 887, 890, 892, 895, 897, 898, 899, 900, 902, 903, 904, 905, 906, 907,
> 908, 910, 914, 916, 918, 919, 920, 921, 922, 925, 926, 928, 929, 933, 935,
> 936, 939, 940, 943, 945, 950, 951, 952, 953, 960, 961, 963, 964, 966, 967,
> 972, 973, 975, 977, 979, 980, 982, 983, 984, 985, 987, 989, 991, 992, 995,
> 996, 999, 1002, 1003, 1005, 1007, 1011, 1014, 1015, 1016, 1018, 1020, 1021]
> >> Event Detail:CacheRebalancingEvent [cacheName=ASSET_GROUP_CACHE,
> part=413, discoNode=TcpDiscoveryNode
> [id=1bb17828-3556-499f-a4e6-98cfdc1d11fb, addrs=[0:0:0:0:0:0:0:1,
> 10.113.14.98, 127.0.0.1], sockAddrs=[], discPort=47501, order=2,
> intOrder=2, lastExchangeTime=1568357181089, loc=false,
> ver=2.6.0#20180710-sha1:669feacc, isClient=false], discoEvtType=12,
> discoTs=1568357376683, discoEvtName=NODE_FAILED, nodeId8=499400ac,
> msg=Cache rebalancing event., type=CACHE_REBALANCE_PART_DATA_LOST,
> tstamp=1568357423500]
> >> primary partition size after loss:1024
> >> Lost partion No.524
> >>
> >> The number of lost partitions  gets increment on each consecutive event
> call and the last many calls to listener have complete list of lost
> partitions.
> >> Questions: 1)Is there any way get list of complete lost partitions?
> Because I want start cache loading for these partitions. Its getting
> difficult to determine when to call cache loading due event call on per
> partition.
> >>                     2)I want reset only partitions using
> resetLostPartitions() those I have handled in partition lost event listener.
> >>
> >> Thanks,
> >> Akash
>

Reply via email to