News

(Update 4/10/2020)[Postpone]ECL2.0 Virtual Network Infrastructure Maintenance


Mar.27th,2020 Published

Apr.10th,2020 Postpone Maintenance Schedule

 

Thank you for using NTT Ltd. Enterprise Cloud 2.0 (ECL) service.

With the spread of new coronavirus infection worldwide, we would like to postpone the maintenance schedule of the following sites.

In our judgement, the impact of the maintenance work on the virtual network infrastructure equipment which we announced the other day will be extremely large to customer.

 

The details are as follows:

Regions:

JP1, JP5, JP2, HK1, SG1, DE1, UK1, US1

Maintenance Schedule:

JP2(Osaka 5th Data Center/Osaka 1st Data Center)

[Postpone] Apr./11/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group A

[Postpone] Apr./18/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group C and Gateway services*1

[Postpone] Apr./25/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group B

[Postpone] May/16/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group D

[Postpone] Jun./20/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Relay systems

 

DE1 (Germany Frankfurt 1 Data Center)

[Postpone] Apr./11/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group A and Gateway services*1

[Postpone] Apr./18/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group D

[Postpone] Apr./25/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group B

[Postpone] May/16/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group C

[Postpone] Jun./20/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Relay systems

 

US1 (Virginia Ashburn1 (VA1) Data Center)

[Postpone] Apr./11/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group A and Gateway services*1

[Postpone] Apr./18/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group A

[Postpone] Apr./25/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group B

[Postpone] Jun./20/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Relay systems

 

HK1 (Hong Kong Financial Data Center)

[Postpone] Apr./11/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group B

[Postpone] Apr./18/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Gateway services*1

[Postpone] May/16/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group A

[Postpone] Jun./20/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :   Relay systems

 

JP1 (Saitama No.1 Data Center)

[Postpone] May/23/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group A

[Postpone] May/30/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group B

[Postpone] Jun./6/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Gateway services*1

[Postpone] Jun./27/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Relay systems

 

JP5 (Tokyo No.2 Data Center)

[Postpone] May/23/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group A

[Postpone] May/30/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group B

[Postpone] Jun./13/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Gateway services*1

[Postpone] Jun./27/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Relay systems

 

UK1 (UK Hemel Hempstead2 / UK Hemel Hempstead3)

[Postpone] May/30/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group A

[Postpone] Jun./6/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group B

[Postpone] Jun./13/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :   Gateway services*1

[Postpone] Jun./27/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Relay systems

 

SG1(Singapore Serangoon Data Center)

[Postpone] May/23/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group B

[Postpone] Jun./6/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Group A

[Postpone] Jun./13/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :    Gateway services*1

[Postpone] Jun./27/2020 (Sat.) 00:00 (UTC) ~ 15:00 (UTC)  :  Relay systems

 

*1: Internet Connectivity, VPN Gateway, Amazon Web Services Inter-Connectivity, Google Cloud Platform Inter-Connectivity, Microsoft Azure Inter-connectivity, DC Inter-Connectivity, Colocation Inter-Connectivity (CIC)(1000BASE-T) , Colocation Inter-Connectivity (CIC)(1Gbps(1000BASE-LX)/10Gbps(10GBASE-LR)), Enterprise Cloud 1.0 Inter-Connectivity (EIC)

 

Impacted Services during maintenance window

  1. Impacted service (Server and Network Service )

There will be no impact on service currently in use, such as Virtual Server interruption, disconnection of network service during maintenance.

Guest OS and applications on the virtual servers (ex. Connected by SSH) will be available without interruption.

 

  1. GUI (ECL2.0 portal) and APIs

(1) Operational impact on ECL2.0 portal and API

The following operations of Portal / API may become unstable during maintenance of each target region/ group.

Details are as follows.

– Operations of Portal and API

– Upload/Download image (During maintenance in Group A)

– Create volume and Delete volume

– Getting backup

 

(2) Service impacts during maintenance

Service impacts vary by the maintenance in each region (Group A / B / C / D, gateway services, relay systems) and affected services.

  1. a) Group A/B/C/D

Impacted Services

Virtual Server, Firewall(Brocade 5600 vRouter) , Firewall(vSRX), Load Balancer(NetScaler VPX),

Managed Firewall, Managed UTM, Managed WAF, Baremetal Server, Block Storage (Provisioned I/O Performance), FileStorage

 

Service Impacts

Impacted Services 1:

Virtual Server, Firewall(Brocade 5600 vRouter) , Firewall(vSRX), Load Balancer(NetScaler VPX),

Managed Firewall, Managed UTM, Managed WAF, Baremetal Server(Remote Console Access (RCA))

Service impacts : Up to 3 sec. x few times

 

Impacted Services 2:

Block Storage (Provisioned I/O Performance), FileStorage

Service impacts :  Up to 60 sec x  few times

 

Impacted Services 3  : Baremetal Server

Service impacts:  There will be approximately 30 minutes’ communication interruptions for a single system in each plane (Data, Storage) .

 

* There might be a instantaneous communication interruption or no communication interruption at all if you have a redundant configuration.

Baremetal Server Network Settings

※ Depending on the maintenance situation, communication interruptions for the virtual server and communication instability may occur up to 30 minutes.(According to the verification results in the our lab, the incidence rate was 0.3%)

 

 

b)Gateway Services

Impacted Services

Internet Connectivity, VPN Gateway, Amazon Web Services Inter-Connectivity,  Google Cloud Platform Inter-Connectivity,

Microsoft Azure Inter-connectivity, DC Inter-Connectivity, Colocation Inter-Connectivity (CIC), Enterprise Cloud 1.0 Inter-Connectivity (EIC)

 

Service Impacts

Impacted Services1: Internet Connectivity, VPN Gateway, Amazon Web Services Inter-Connectivity,

Google Cloud Platform Inter-Connectivity, Microsoft Azure Inter-connectivity, DC Inter-Connectivity,

Colocation Inter-Connectivity (CIC)(1000BASE-T), Enterprise Cloud 1.0 Inter-Connectivity (EIC)

Service impacts: Up to 60 sec x a few times

 

Impacted Services1: Colocation Inter-Connectivity (CIC)(1Gbps(1000BASE-LX)/10Gbps(10GBASE-LR))

Service impacts:

There will be approximately 30 minutes’communication interruption for each physical connection.

* There might be a instantaneous communication interruption or no communication interruption at all if you have a redundant configuration.

 

Colocation Inter-Connectivity (CIC) – Restrictions – Redundancy Configurations and Availability

https://ecl.ntt.com/en/documents/service-descriptions/rsts/network/cic.html

※ Depending on the maintenance situation, communication interruption for gateway services and communication instability may occur up to 30 minutes (according to laboratory tests, the incidence rate was 0.3%)

 

 

c)  Relay systems

Impacted Services:  All services in the maintenance region

Service Impacts:  Up to 3 sec. x 4 times

FAQ

No.1

Q:

Is there a way to avoid the maintenance impact?

 

A:

If there is an inter-region redundancy configuration or an inter-group redundancy configuration, the effect of maintenance can be avoided by switching regions, and the effect of maintenance can be reduced by switching between inter-group redundancy.

 

Regarding the maintenance of group A / B / C / D, if there is redundancy between groups,switching from a maintenance target group to another redundancy group does not affect service.

 

Since the gateway service does not belong to a group, the maintenance of the gateway service cannot avoid the effects on switching redundancy between groups. If you have an inter-region redundancy configuration, you can avoid the impact of maintenance by switching between inter-region redundancy.

 

Switching operation between regions and groups depends on the customer system. Check with IT department or construction vendor. In addition, there may be an impact when switching the system depending on the application,Therefore check the effects of the system including the application before switching the system.

 

 

No.2

Q:

Are there any operations required from customer side before and after the maintenance?

A:

Let customer confirm the settings of the below items and change the settings if necessary(i)Redundant configuration of Baremetal Server, Colocation Inter-Connectivity (CIC)(1Gbps(1000BASE-LX)/10Gbps(10GBASE-LR)),  (ii) Redundant configuration of Firewall(Brocade 5600 vRouter), Firewall(vSRX),  Load Balancer(NetScaler VPX), Managed Firewall, Managed UTM and (iii) Connection of Virtual Server and Block Storage (Provisioned I/O Performance), FileStorage.

 

(i) About Baremetal Server, Colocation Inter-Connectivity (CIC)(1000Base-LX / 10GBase-LR) Redundancy Configuration

 

Customers who are using Baremetal Server, Colocation Inter-Connectivity (CIC) (1000Base-LX / 10GBase-LR) Below impacts may occur during the maitenance.

  • Baremetal Server

There will be approximately 30 minutes’ communication interruptions for a single system in each plane (Data, Storage) .

  • Colocation Inter-Connectivity (CIC) (1000Base-LX / 10GBase-LR)

There may be approximately 30 minutes’s communication interruption for each physical connection happening one after another.

In the case of a redundant configuration, the effects of maintenance can be reduced by switching the system.

 

We recommend that you check your redundancy settings before the maintenance.

Please, refer to the (2) Service impacts during maintenance.

 

(ii) About Firewall(Brocade 5600 vRouter) , Firewall(vSRX), Load Balancer(NetScaler VPX), Managed Firewall, Managed UTM Redundancy Configuration

 

Switch between active and redundant side of Firewall(Brocade 5600 vRouter) , Firewall(vSRX), Load Balancer(NetScaler VPX), Managed Firewall and  Managed UTM may occur during the maintenance. We recommend customers confirm the redundant configuration (VRRP related and DHCP settings in logical network) before the operation.

Please refar to below for details.

 

Managed Firewall/UTM

https://ecl.ntt.com/documents/tutorials/security/rsts/security/operation/managed_firewall_utm/3120_interface_ha.html

https://ecl.ntt.com/documents/tutorials/security/rsts/security/operation/managed_firewall_utm/8080_ng_setting_guide.html?

Load Balancer(NetScaler VPX)

https://ecl.ntt.com/documents/tutorials/rsts/LoadBalancer/network/lb_vrrp.html?

Firewall(Brocade 5600 vRouter)

https://ecl.ntt.com/documents/tutorials/rsts/Firewall/vyatta/command.html?

Firewall(vSRX)

https://ecl.ntt.com/documents/tutorials/rsts/vSRX/network/vrrp.html#vrrp-vrrp

 

Logical Network

https://ecl.ntt.com/documents/tutorials/rsts/Network/logicalnw.html

Known Issues

https://ecl.ntt.com/known-issues/vrrp-configuration/?

 

 

(iii) About connection settings between virtual server and block storage

Regarding the traffic between virtual server and block storage, unicast interruption my occur on virtual servers. However, if the customer use official template and set configuration according to the tutorial, iSCSI coonection interruption may happen but disk I/O communication will not be interrupted because of the multi-path device functionality.However there might be some intterutions depending on the applications hence please check the configuration of each tenant by customer.

Also we highly recommend that customer who don’t set multi-path configuration change the configuration of time-out amount of iSCSI initiator as “120 seconds”. (Please check impact on each application when changing configurations.)

 

 

No.3

Q:

Can I opertate the server during maitenance when necessary?

 

A:

Operation using the portal and API is possible even during maintenance.

However, depending on the status of the maintenance, delays and processing failure may occur. We recommend you arrange the operation schedule to another day apart from maitenance period.

 

 

No.4

Q:

Please describe the impact on the application layer due to the network service impact caused by the maintenance.

 

A:

We don’t know how it affects your application.

The effect on the application due to the communication effect caused by maintenance work depends on the customer system.

Please contact the system department or contractor that is responsible for building the customer’s system.

 

 

No.5

Q:

Which menu especially affects unicast communication?

A:

Please refer to  (2) Service impacts during maintenance.

 

 

No.6

Q:

It is said that communication disruption and on virtual server and I/O instability may occur up to 30 minutes depending on maintenance situation the notification email. What is the possibility?

A:

According to lab test results, the incidence rate was 0.3%.

If it happens just in case, we will take remedial action immediately.