Deployement failed step 31 of 32 :Cannot connect to Virtual Controller after reboot

[RicT1]

Hi,

I tried to deploy a single Simplivity DL380 Gen 10 XS node in version 3.7.6 and each time i have the following error: step 31 of 32 -Cannot connect to Virtual Controller after rebooting it.

After the installation error, i can connect normally to the ESXi (10.0.0.116) and the OVC (10.0.0.117).

I found this error on orchestrator log file:

2019-01-30 14:45:29,985Z INFO  OmniStack-10.0.0.116 [c.s.d.o.Orchestrator] connect(SvtControlConnect.java:82) - Connection to OmniCube at 10.0.0.117 port 9190 failed - exception java.net.ConnectException: Connection refused: connect
java.net.ConnectException: Connection refused: connect
    at java.net.DualStackPlainSocketImpl.waitForConnect(Native Method)
    at java.net.DualStackPlainSocketImpl.socketConnect(Unknown Source)
    at java.net.AbstractPlainSocketImpl.doConnect(Unknown Source)
    at java.net.AbstractPlainSocketImpl.connectToAddress(Unknown Source)
    at java.net.AbstractPlainSocketImpl.connect(Unknown Source)
    at java.net.PlainSocketImpl.connect(Unknown Source)
    at java.net.SocksSocketImpl.connect(Unknown Source)
    at java.net.Socket.connect(Unknown Source)
    at sun.security.ssl.SSLSocketImpl.connect(Unknown Source)
    at com.simplivity.util.network.AutoCert.initSSLSocket(AutoCert.java:224)
    at com.simplivity.util.network.AutoCert.getSvtTrustStore(AutoCert.java:191)
    at com.simplivity.util.network.AutoCert.verifyCert(AutoCert.java:110)
    at com.simplivity.util.network.AutoCertThrift.getTransport(AutoCertThrift.java:51)
    at com.simplivity.deploy.orchestrator.SvtControlConnect.connect(SvtControlConnect.java:73)
    at com.simplivity.deploy.orchestrator.DeployActions.connectSVT(DeployActions.java:3319)
    at com.simplivity.deploy.orchestrator.CustomerDeploy.runSteps(CustomerDeploy.java:163)
    at com.simplivity.deploy.orchestrator.DeployExecutor.run(DeployExecutor.java:439)
    at java.lang.Thread.run(Unknown Source)
2019-01-30 14:45:29,986Z ERROR OmniStack-10.0.0.116 [c.s.d.o.Orchestrator] error(DeployLogger.java:98) - step 31 of 32 (49.78%) - Cannot connect to Virtual Controller after rebooting it - at 10.0.0.117

Thanks for help on this problem

Regards

[DeclanOR]

Hi @RicT1

Thanks for your query.

You will probably need to raise a support ticket to look into this with you, however, can you maybe try one thing?

Can you power down the node for 5 mins, and then re-attempt the deployment? If this doesn't help, then please raise a ticket.

Thanks,

Declan.

[RicT1 · Accepted solution]

Hi,

The issue was cause by a faulty GPU card. The CPU card was blinking red at the rear of the server. HPe support will replace the GPU card.

Also, i could connect the OVC after the failed deployement and i found the following error with dmesg command.

[ 56.498229] IA0: C2S Console: [ 6169.240000] xi2cps e0005000.ps7-i2c: I2C_ERROR err_status 0x4 isr 0x4 CR 0x320e SR 0x40 recv_count 0 send_count 0
[ 56.508455] IA0: C2S Console: [ 6169.250000] envmon ERROR: get_i2c_register:418 Unable to send I2c data: Input/output error
[ 56.517019] IA0: C2S Console: [ 6169.260000] envmon ERROR: get_STTS2002_temp:444 Unable to get I2C register! bus /dev/i2c-1, addr 0x18 reg 0x0: Input/output error
[ 56.529014] IA0: C2S Console: [ 6169.270000] xi2cps e0005000.ps7-i2c: I2C_ERROR err_status 0x4 isr 0x4 CR 0x320e SR 0x40 recv_count 0 send_count 0
[ 56.539241] IA0: C2S Console: [ 6169.280000] envmon ERROR: get_i2c_register:418 Unable to send I2c data: Input/output error
[ 56.547745] IA0: C2S Console: [ 6169.290000] envmon ERROR: get_STTS2002_temp:444 Unable to get I2C register! bus /dev/i2c-1, addr 0x18 reg 0x0: Input/output error

Thanks,

Ric

[DeclanOR]

@RicT1 

Thanks for confirming. I suspect you mean the TIA (accelerator) card correct?

This is what I suspected. A drain of the flea power as suggested often resolves issues with the TIA card which is why I recommended the power down for 5 minutes. If the TIA was actually faulty, then this would not have helped either.

Glad to hear you got this resolved with help of support.

Best ragrds,

DeclanOR