backblaze b2 / amazon s3 as remote in xoa

shorian

Update to my earlier post - We found the connection timeout issue was solved by allocating more memory to XO. Even though above I said that the memory didn't appear to be a problem, it turned out that Debian was swapping out so as to keep a chunk of free memory available, so we mistakenly assumed that not using all the memory meant we had sufficient. However being memory restricted combined with a slow disk meant that the swap was growing faster than it was being processed.

Substantially increasing the XO VM memory (4->16gb) seems to have solved the timeout issues (so yes, root cause was user error), and we're now finding that the S3 api to B2 (a lot cheaper than Amazon) is working really well for us.

Well done to the XO dev team - the S3 api has completely changed how we use backups and freed up a lot infrastructure that we previously had had to dedicate to this; thank you

olivierlambert

@nraynaud will be happy to hear that

nraynaud

@shorian Thank you.

I will look into the memory issue, that seems to be the clue into a hidden issue, thank you.

nraynaud

@shorian I think I found the memory hog in S3 thanks to your message: https://github.com/vatesfr/xen-orchestra/pull/5456

nraynaud opened this pull request in vatesfr/xen-orchestra

closed WiP: fix(backup/s3): maybe fix S3 timeout #5456

julien-f

@akurzawa @DodgeThis @shorian Any of you able to test this branch to see whether it fixes the issue?

DodgeThis

@julien-f No problem. glad to test it.

DodgeThis

@julien-f
Still no sucess,
This vm that i'm trying to backups is about 290 Gb

Xen Orchestra is on a VM with:
Debian 10
4 Gb Ram
85 GB free hd
1CPU

xo-server 5.72.1
xo-web 5.75.0

{
  "data": {
    "mode": "full",
    "reportWhen": "failure"
  },
  "id": "1608035511007",
  "jobId": "695f3dfb-783f-4cb9-85ef-6805d75c7d26",
  "jobName": "AmazonTest",
  "message": "backup",
  "scheduleId": "75c619c3-8b6d-4c72-b110-131b01992b3e",
  "start": 1608035511007,
  "status": "failure",
  "tasks": [
    {
      "data": {
        "type": "VM",
        "id": "93b88999-8b61-7207-138b-1ef389491e71"
      },
      "id": "1608035511027",
      "message": "Starting backup of MysqlDB. (695f3dfb-783f-4cb9-85ef-6805d75c7d26)",
      "start": 1608035511027,
      "status": "failure",
      "tasks": [
        {
          "id": "1608035511031",
          "message": "snapshot",
          "start": 1608035511031,
          "status": "success",
          "end": 1608035512857,
          "result": "e612bb3b-9141-45ef-a901-ac8e4f1927c2"
        },
        {
          "id": "1608035512859",
          "message": "add metadata to snapshot",
          "start": 1608035512859,
          "status": "success",
          "end": 1608035512869
        },
        {
          "id": "1608035513055",
          "message": "waiting for uptodate snapshot record",
          "start": 1608035513055,
          "status": "success",
          "end": 1608035513259
        },
        {
          "id": "1608035513261",
          "message": "start VM export",
          "start": 1608035513261,
          "status": "success",
          "end": 1608035513302
        },
        {
          "data": {
            "id": "72341e61-56db-43f5-8389-78baf364516f",
            "type": "remote"
          },
          "id": "1608035513304",
          "message": "export",
          "start": 1608035513304,
          "status": "failure",
          "tasks": [
            {
              "id": "1608035514389",
              "message": "transfer",
              "start": 1608035514389,
              "status": "failure",
              "end": 1608035839218,
              "result": {
                "message": "write EPIPE",
                "errno": "EPIPE",
                "code": "NetworkingError",
                "syscall": "write",
                "region": "eu-central-1",
                "hostname": "s3.eu-central-1.amazonaws.com",
                "retryable": true,
                "time": "2020-12-15T12:37:18.736Z",
                "statusCode": 400,
                "name": "NetworkingError",
                "stack": "Error: write EPIPE\n    at WriteWrap.onWriteComplete [as oncomplete] (internal/stream_base_commons.js:94:16)"
              }
            }
          ],
          "end": 1608035839357,
          "result": {
            "message": "write EPIPE",
            "errno": "EPIPE",
            "code": "NetworkingError",
            "syscall": "write",
            "region": "eu-central-1",
            "hostname": "s3.eu-central-1.amazonaws.com",
            "retryable": true,
            "time": "2020-12-15T12:37:18.736Z",
            "statusCode": 400,
            "name": "NetworkingError",
            "stack": "Error: write EPIPE\n    at WriteWrap.onWriteComplete [as oncomplete] (internal/stream_base_commons.js:94:16)"
          }
        },
        {
          "id": "1608035839607",
          "message": "set snapshot.other_config[xo:backup:exported]",
          "start": 1608035839607,
          "status": "success",
          "end": 1608035839831
        }
      ],
      "end": 1608039896381
    }
  ],
  "end": 1608039896383
}

DodgeThis

Another one with a VM with +/-600 Gb

{
  "data": {
    "mode": "full",
    "reportWhen": "failure"
  },
  "id": "1608043681979",
  "jobId": "abfed8aa-b445-4c94-be5c-b6ed723792b8",
  "jobName": "amazontest",
  "message": "backup",
  "scheduleId": "44840d13-2302-45ae-9460-53ed451183ed",
  "start": 1608043681979,
  "status": "failure",
  "infos": [
    {
      "data": {
        "vms": [
          "fbea839a-72ac-ca3d-51e5-d4248e363d72"
        ]
      },
      "message": "vms"
    }
  ],
  "tasks": [
    {
      "data": {
        "type": "VM",
        "id": "fbea839a-72ac-ca3d-51e5-d4248e363d72"
      },
      "id": "1608043681986:0",
      "message": "Starting backup of DESKTOPGREEN. (abfed8aa-b445-4c94-be5c-b6ed723792b8)",
      "start": 1608043681986,
      "status": "failure",
      "tasks": [
        {
          "id": "1608043681990",
          "message": "snapshot",
          "start": 1608043681990,
          "status": "success",
          "end": 1608043683927,
          "result": "42f38a1a-5037-107d-1ceb-921fed6910fa"
        },
        {
          "id": "1608043683929",
          "message": "add metadata to snapshot",
          "start": 1608043683929,
          "status": "success",
          "end": 1608043683938
        },
        {
          "id": "1608043684127",
          "message": "waiting for uptodate snapshot record",
          "start": 1608043684127,
          "status": "success",
          "end": 1608043684330
        },
        {
          "id": "1608043684332",
          "message": "start VM export",
          "start": 1608043684332,
          "status": "success",
          "end": 1608043684348
        },
        {
          "data": {
            "id": "7847ac06-d46f-476a-84c1-608f9302cb67",
            "type": "remote"
          },
          "id": "1608043684352",
          "message": "export",
          "start": 1608043684352,
          "status": "failure",
          "tasks": [
            {
              "id": "1608043684783",
              "message": "transfer",
              "start": 1608043684783,
              "status": "failure",
              "end": 1608043987240,
              "result": {
                "message": "Your socket connection to the server was not read from or written to within the timeout period. Idle connections will be closed.",
                "code": "RequestTimeout",
                "region": null,
                "time": "2020-12-15T14:53:06.821Z",
                "requestId": "E47562F2C7A82C45",
                "extendedRequestId": "5nM1888rfrbLHHgUrlsuF5cISTd2kj6oVpWfz5Yhtpj4GtfW7wkX9HnTq/Tj8uU37Opp8FNjlFU=",
                "statusCode": 400,
                "retryable": true,
                "name": "RequestTimeout",
                "stack": "RequestTimeout: Your socket connection to the server was not read from or written to within the timeout period. Idle connections will be closed.\n    at Request.extractError (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/services/s3.js:700:35)\n    at Request.callListeners (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:106:20)\n    at Request.emit (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:78:10)\n    at Request.emit (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:688:14)\n    at Request.transition (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:22:10)\n    at AcceptorStateMachine.runTo (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/state_machine.js:14:12)\n    at /opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/state_machine.js:26:10\n    at Request.<anonymous> (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:38:9)\n    at Request.<anonymous> (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:690:12)\n    at Request.callListeners (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:116:18)\n    at Request.emit (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:78:10)\n    at Request.emit (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:688:14)\n    at Request.transition (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:22:10)\n    at AcceptorStateMachine.runTo (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/state_machine.js:14:12)\n    at /opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/state_machine.js:26:10\n    at Request.<anonymous> (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:38:9)\n    at Request.<anonymous> (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:690:12)\n    at Request.callListeners (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:116:18)\n    at callNextListener (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:96:12)\n    at IncomingMessage.onEnd (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/event_listeners.js:313:13)\n    at IncomingMessage.apply (events.js:326:22)\n    at IncomingMessage.patchedEmit (/opt/xo/xo-builds/xen-orchestra-202012150908/@xen-orchestra/log/src/configure.js:92:17)"
              }
            }
          ],
          "end": 1608043987241,
          "result": {
            "message": "Your socket connection to the server was not read from or written to within the timeout period. Idle connections will be closed.",
            "code": "RequestTimeout",
            "region": null,
            "time": "2020-12-15T14:53:06.821Z",
            "requestId": "E47562F2C7A82C45",
            "extendedRequestId": "5nM1888rfrbLHHgUrlsuF5cISTd2kj6oVpWfz5Yhtpj4GtfW7wkX9HnTq/Tj8uU37Opp8FNjlFU=",
            "statusCode": 400,
            "retryable": true,
            "name": "RequestTimeout",
            "stack": "RequestTimeout: Your socket connection to the server was not read from or written to within the timeout period. Idle connections will be closed.\n    at Request.extractError (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/services/s3.js:700:35)\n    at Request.callListeners (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:106:20)\n    at Request.emit (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:78:10)\n    at Request.emit (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:688:14)\n    at Request.transition (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:22:10)\n    at AcceptorStateMachine.runTo (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/state_machine.js:14:12)\n    at /opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/state_machine.js:26:10\n    at Request.<anonymous> (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:38:9)\n    at Request.<anonymous> (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:690:12)\n    at Request.callListeners (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:116:18)\n    at Request.emit (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:78:10)\n    at Request.emit (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:688:14)\n    at Request.transition (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:22:10)\n    at AcceptorStateMachine.runTo (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/state_machine.js:14:12)\n    at /opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/state_machine.js:26:10\n    at Request.<anonymous> (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:38:9)\n    at Request.<anonymous> (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/request.js:690:12)\n    at Request.callListeners (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:116:18)\n    at callNextListener (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/sequential_executor.js:96:12)\n    at IncomingMessage.onEnd (/opt/xo/xo-builds/xen-orchestra-202012150908/node_modules/aws-sdk/lib/event_listeners.js:313:13)\n    at IncomingMessage.apply (events.js:326:22)\n    at IncomingMessage.patchedEmit (/opt/xo/xo-builds/xen-orchestra-202012150908/@xen-orchestra/log/src/configure.js:92:17)"
          }
        },
        {
          "id": "1608043987670",
          "message": "set snapshot.other_config[xo:backup:exported]",
          "start": 1608043987670,
          "status": "success",
          "end": 1608043989451
        }
      ],
      "end": 1608051207208
    }
  ],
  "end": 1608051207238
}

DodgeThis

@julien-f shorian like @shorian says increasing mem to 16 Gb fixed the problem

julien-f

@nraynaud ping

nraynaud

@dodgethis thank you, I will look harder into this.

DodgeThis

@nraynaud glad to help somehow

klou

@nraynaud

Confirmed bumping to 12 Gb RAM allowed my "larger" VM's (60 GB+) to complete.

xo-server 5.73.0
xo-web 5.76.0

nraynaud

Can someone please test the issue with the current master branch please?

shorian

@nraynaud Will aim to do this over the weekend

shorian

I cannot get the backup to run; I have now tried about 12 times to run a backup that should complete in under 2 hours; without a timeout it runs on (stopped it after 10 hours), with a timeout of 3 hours it consistently fails with 'interrupted' without explanation.

Looking through the logs, we've got a lot of buffer allocation errors, it starts with this:

xo:xo-server ERROR uncaught exception {
  error: Error [RangeError]: Array buffer allocation failed
      at new ArrayBuffer (<anonymous>)
      at new Uint8Array (<anonymous>)
      at new FastBuffer (internal/buffer.js:950:1)
      at createUnsafeBuffer (buffer.js:149:12)
      at allocate (buffer.js:419:10)
      at Function.allocUnsafe (buffer.js:384:10)
      at BufferList._getBuffer (/opt/xen-orchestra/@xen-orchestra/fs/node_modules/readable-stream/lib/internal/streams/buffer_list.js:166:24)
      at BufferList.consume (/opt/xen-orchestra/@xen-orchestra/fs/node_modules/readable-stream/lib/internal/streams/buffer_list.js:119:54)
      at fromList (/opt/xen-orchestra/@xen-orchestra/fs/node_modules/readable-stream/lib/_stream_readable.js:1073:24)
      at Transform.Readable.read (/opt/xen-orchestra/@xen-orchestra/fs/node_modules/readable-stream/lib/_stream_readable.js:465:20)
      at ManagedUpload.fillStream (/opt/xen-orchestra/node_modules/aws-sdk/lib/s3/managed_upload.js:422:25)
      at Response.<anonymous> (/opt/xen-orchestra/node_modules/aws-sdk/lib/s3/managed_upload.js:574:24)
      at Request.<anonymous> (/opt/xen-orchestra/node_modules/aws-sdk/lib/request.js:369:18)
      at Request.callListeners (/opt/xen-orchestra/node_modules/aws-sdk/lib/sequential_executor.js:106:20)
      at Request.emit (/opt/xen-orchestra/node_modules/aws-sdk/lib/sequential_executor.js:78:10)
      at Request.emit (/opt/xen-orchestra/node_modules/aws-sdk/lib/request.js:688:14)
      at Request.transition (/opt/xen-orchestra/node_modules/aws-sdk/lib/request.js:22:10)
      at AcceptorStateMachine.runTo (/opt/xen-orchestra/node_modules/aws-sdk/lib/state_machine.js:14:12)
      at /opt/xen-orchestra/node_modules/aws-sdk/lib/state_machine.js:26:10
      at Request.<anonymous> (/opt/xen-orchestra/node_modules/aws-sdk/lib/request.js:38:9)
      at Request.<anonymous> (/opt/xen-orchestra/node_modules/aws-sdk/lib/request.js:690:12)
      at Request.callListeners (/opt/xen-orchestra/node_modules/aws-sdk/lib/sequential_executor.js:116:18) {
    code: 'RangeError',
    time: 2021-01-16T22:07:11.590Z
  }
}

...followed by this repeating numerous times:

xo:xo-server ERROR uncaught exception {
  error: RangeError: Array buffer allocation failed
      at new ArrayBuffer (<anonymous>)
      at new Uint8Array (<anonymous>)
      at new FastBuffer (internal/buffer.js:950:1)
      at createUnsafeBuffer (buffer.js:149:12)
      at allocate (buffer.js:419:10)
      at Function.allocUnsafe (buffer.js:384:10)
      at Function.concat (buffer.js:564:25)
      at ManagedUpload.fillStream (/opt/xen-orchestra/node_modules/aws-sdk/lib/s3/managed_upload.js:433:38)
      at Transform.<anonymous> (/opt/xen-orchestra/node_modules/aws-sdk/lib/s3/managed_upload.js:188:44)
      at Transform.apply (events.js:314:20)
      at Transform.patchedEmit (/opt/xen-orchestra/@xen-orchestra/log/src/configure.js:92:17)
      at Transform.EventEmitter.emit (domain.js:483:12)
      at emitReadable_ (/opt/xen-orchestra/@xen-orchestra/fs/node_modules/readable-stream/lib/_stream_readable.js:538:12)
      at processTicksAndRejections (internal/process/task_queues.js:83:21)
}

Going to try whacking the memory right up ; will report back later today.

shorian

Ok, been running this all weekend. In positive news, I've not had the timeout error once. In less positive news, I've yet to get a successful backup

With lots of memory and unlimited time, I'm afraid we consistently fail with HANDLE_INVALID. I'm unsure where to go with things from here - @nraynaud let me know what I can do to help rather than throw problems at you. I'm at your disposal.

nraynaud

@shorian thank you for your help. can you tell me what kind of backup you were launching and at what phase the error happened ?

shorian

@nraynaud Normal backup to B2, concurrency set to 1, compression set to zsd - no other settings (ie no 'delete before' or 'with memory' - just a standard backup with zstd compression). I have also tried without compression without any variation in outcome.

Error occurs immediately after snapshot (about 5 mins later, but its the next line in the debug log), but we don't fail out for another 18 minutes; we get the bad allocation error every second until we finally die with a terminal 'bad allocation' error and completely fall over with an auto-restart of the script:

2021-01-16T22:20:00.164Z xo:xo-server ERROR uncaught exception {
  error: RangeError: Array buffer allocation failed
      at new ArrayBuffer (<anonymous>)
      at new Uint8Array (<anonymous>)
      at new FastBuffer (internal/buffer.js:950:1)
      at createUnsafeBuffer (buffer.js:149:12)
      at allocate (buffer.js:419:10)
      at Function.allocUnsafe (buffer.js:384:10)
      at Function.concat (buffer.js:564:25)
      at ManagedUpload.fillStream (/opt/xen-orchestra/node_modules/aws-sdk/lib/s3/managed_upload.js:433:38)
      at Transform.<anonymous> (/opt/xen-orchestra/node_modules/aws-sdk/lib/s3/managed_upload.js:188:44)
      at Transform.apply (events.js:314:20)
      at Transform.patchedEmit (/opt/xen-orchestra/@xen-orchestra/log/src/configure.js:92:17)
      at Transform.EventEmitter.emit (domain.js:483:12)
      at emitReadable_ (/opt/xen-orchestra/@xen-orchestra/fs/node_modules/readable-stream/lib/_stream_readable.js:538:12)
      at processTicksAndRejections (internal/process/task_queues.js:83:21)
}
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc
error: Forever detected script was killed by signal: SIGABRT
error: Script restart attempt #2
2021-01-16T22:20:17.950Z app-conf /opt/xen-orchestra/packages/xo-server/config.toml
2021-01-16T22:20:17.956Z app-conf /root/.config/xo-server/config.toml
2021-01-16T22:20:17.957Z app-conf /root/.config/xo-server/config.z-auto.json
2021-01-16T22:20:17.960Z xo:main INFO Configuration loaded.
2021-01-16T22:20:17.963Z xo:main INFO Web server listening on http://[::]:80

Interestingly I've left the backup process to keep running every few hours, and we're now getting some success - I've yet to have the 3 VMs in question all back up in one pass, but the last couple of passes have resulted in one or two of them completing successfully. See sample logs:

2021-01-18T09_00_00.011Z - backup NG.log.txt

2021-01-18T03_00_00.009Z - backup NG.log.txt

In summary, we are now we are getting some timeouts - different than the 12,000ms that we were seeing originally, but now with the new errors in the attached logs of 'no object with UUID or opaque ref'.

If I lower the memory back down to 4gb we revert back to the buffer allocation errors.

I'm very happy to send over more extensive logs but would prefer to do so 1:1 rather than into a public forum.

nraynaud

@shorian thank you I will look at all that.