| Question |
| Difference between Clariion & VNX ? |
| Answer |
|
| One major difference is the architecture of the CLARiiON is totally different from VNX. Apart from this, CLARiiON is based on traditional RAID Groups along with small Storage pool concepts with tiering and VNX is totally based on Storage Pools. However, it is also compatible with traditional RIAD Groups. |
|
| Question |
| What is the use of FAST Cache AND FAST VP ? |
| Answer |
|
| Fast Cache: Enablers flash drives to be used to extend the missing caching capactity of the storage system. Fast VP: Leverages pools to provide sub-LUN tiering, enabling the utilization of multiple tiers of storage simultaneously. Both are designed to improve performance of the array. |
|
| Question |
| How to deal when vault disk has many soft errors? |
| Answer |
| You have to treat the vault drive as any other normal drive which has soft errors on it. If there are two vault drives which are reporting soft errors and also there are two vault drives vault in the array then you need to log a ticket with EMC and EMC support team will engage internal Engineering team. |
|
| Question |
| Procedure to add an enclosure in running environment ? |
| Answer |
| You need to connect the expansion cable from the last enclosure in the array to the new enclosure and configure the Bus and Enc ID. It will be automatically detected by the FLARE and then you can add new drives to the enclosure. |
|
| Question |
| What is basic difference failover between DataMover and SP ? |
| Answer |
|
| DataMover is Celerra component which serves I/Os to the connected host over the TCP/IP network. SP is Storage Process which is the main component of CLARiiON and VNX block storage which servies the I/Os to the connected hosts over Fibre and It also manages the other storage functions like vaulting, LUNs ownership. Operating system for DataMover is DART Operating system of SP is FLARE. |
|
| Question |
|
| Any best suitable website is there for getting knowledge other than powerlink? |
| Answer |
| Powerlink is outdated. You can always use the new support website of EMC: https://support.emc.com Apart from this, you can also refer to http://community.emc.com There are EMC experts on EMC community website, they answers questions and also write blogs on different issues and concepts. |
|
| Question |
| Brief us about the EMC CLARiiON ? |
|
| Answer |
| EMC CLARiiON is a block level mid range storage and here we are covering the architecture of the CLARiiON and basic troubleshootings steps that an admin can follow before loggin a ticket with EMC support team. |
|
| Question |
|
| If we use vault drives is there any issue for valting? |
| Answer |
|
| No there is no issue with vaulting if you use vault drives to bound user LUNs but the performance of the array will be impacted. The best practice according to the EMC is not to use vault drives for any users LUNs but you need to bind one test LUN on vault drives so that if the any of the vault drive is getting failed (0.0.0 to 0.0.4), user will get notification via call home template. |
|
| Question |
| If we want to use vault drives space for data, How much space we can use? |
| Answer |
| It is not recommended to use vault drives for user LUNs as it will impact the performance of the array but you can use it at your own risk and the usable space depends upon the drive size of the vault drives. The best practice according to the EMC is not to use vault drives for any users LUNs but you need to bind one test LUN on vault drives so that if the any of the vault drive is getting failed (0.0.0 to 0.0.4), user will get notification via call home template. |
|
| Question |
| What is the basic difference between Storage POOL and RAID Group? |
| Answer |
| Storage Pool is the mix of drives and RAID Group can be configured with only one drive types. In one Raid group, you can only configure max of 16 drives. In brief, a storage pool removes the constraint of only 15 max disks you could use in one RAID group. DIsks of different type and performance characteristics can be added in the same pool. If FAST is enabled, then the array does the automatic monitoring of data access patter and moves data across different tiers in the storage pool but RAID grups has no such feature. |
|
| Question |
| What is the procedure to find out the reason for the scenario where one SP is in unmanaged state? |
| Answer |
| You can only check the below things at your level : 1) IP of the SP is pingable or not 2) SP is serving I/Os or not. You can verify this by checking if there is any LUN with the ownership of the SP which is in unmanaged state Apart from this, the unmanaged state of the SP also happens due to fault in NaviCIMOM.exe process which runs on the FLARE. If this process is hung on one SP, you will get unamanaged state of that SP. |
|
| Question |
| Can we mix different types of HDDs in same enclosure ? |
| Answer |
| It depends that what type of drives you are mixing in one enclosure in CLARiiON. If Celerra is used as front end, then you cannot mix drives in one enclosure. You cannot mix FC and SATA drives in one enclosure however you can mix drive types within the bus in different enclosures. |
|
| Question |
| How to identify the flare version in clarion like and How will determine it ? |
| Answer |
| You can verify the FLARE version using the Navisphere or Unisphere GUI by going the storage properties window. Apart from this you can also use NaviCLI commands >> naviseccli -h -user -password -scope <0/1> getagent |
|
| Question |
| If I login with VNX SP A – IP Address,can I manage the Block level and file level management ? |
|
| Answer |
| Yes, If you have VNX unified storage and SP-A is the master domain IP address then you can manage both block and file level management. |
|
| Question |
|
| Is it possible to configure e-mail alert notification for particular SPs on VNX? |
| Answer |
| It is not recommended to do that. Ideally the call home template is configured on both the SP’s so that if one SP is down the other SP can generate call home for all the configured alerts/warnings. |
|
| Question |
| Do we have any benchmark for number of IOPS that could be handled by CX4-120 till 960 series for a standard sized IO of 8 KB |
| Answer |
| There is no benchmark for the number of IOPS that we can handle on CX4-120 till 960. It all depends upon the type of drives we have installed on the box. I/Os that array can handle depends upon the size and type of the drives and also the configuration of RAID. |
|
| Question |
| WHAT IS THE O/S in VNX UNIFIED STOARGE?INCLUDING FILE AND BLOCK LEVEL. |
| Answer |
| Block OS is FLARE File OS is DART and Control station runs on customized linux. |
|
| Question |
| If we have free/empty slots with B0E0 (first enclosure) can we insert new drives? If yes, what are the things we need to consider? Like drive type should be same(EFD/FC/SATA/SATA) |
|
| Answer |
| Yes you can insert new drives but it depends on the below scenarios: * What type of drives are already installed on that enclosure. You cannot mix FC and SATA drives in the same enclosure. There is a separate recommendations for installing and configuring EFD drives in one enclosure. You can reach out to me for any question regarding this @ shivam.chawla@wipro.com |
|
| Question |
| CX-4-480 04.30.000.5.523 cx4 480- 04.30.000.5.524 cx4-120 – 04.29.000.5.003 these are my storage & their firmware versons ,latest is 04.30.000.5.526 does the 2 cx4-480 systems need to update? |
| Answer |
| Yes, You need to upgrade all three boxes to FLARE 04.30.000.5.526 |
|
| Question |
| What basic steps we have to take when storage get down ? lease suggest prior precautions also to avoid storage issues. |
|
| Answer |
| It all depends upon the scenarios why storage is down. It can be due to hardware fault or FLARE fault and there are other software fault also like replication, performance. Apart from this, the Data Unavailability can also happen because of connectivity issue between the storage and the server. Server HBA can also be the cause of environment down. |
|
| Question |
| If I have 2 raid 5 groups 1 is highly important,get 1 disk fail i have no disk, i want to pull adisk from 2nd raid5 which has some not so important data.can i do this ?or i have to format that? if so how can i format? |
| Answer |
| It is not possible to pull out a disk from a RAID 5 Raid Group and use it in the place of faulted drive of another RAID 5 Raid Group. If you want to do that, you have to first remove all the LUNs from the RAID group which is not important to you and then destroy that Raid Group. All the drives of the RAID group which is not important, will become unbounded and then you can use those disks. You can only use unbounded disks to replace any faulted disk. |
|
| Question |
| What is Persistence Mode? |
| Answer |
| Persistence mode will protect the snapshot sessions from getting failed. due to any hardware fault on the storage or LUN trespass. In non persistence mode, If a LUN trespasses to the peer SP,. snap cache may follow the source LUN to the other SP stopping the sessions. In this case, the snapview session was not started in persistent mode, which creates a session that can withstand the following failures and trespasses: * SP reboot or failure * Storage system reboot or power failure * Server I/O trespassing to the peer SP |
|
| Question |
| What is persistence mode? |
| Answer |
| Persistence mode will protect the snapshot sessions from getting failed. due to any hardware fault on the storage or LUN trespass. In non persistence mode, If a LUN trespasses to the peer SP,. snap cache may follow the source LUN to the other SP stopping the sessions. In this case, the snapview session was not started in persistent mode, which creates a session that can withstand the following failures and trespasses: * SP reboot or failure * Storage system reboot or power failure * Server I/O trespassing to the peer SP |
|
| Question |
| Can I expand a raid5 with 5 drives with another raid5 with 3 same/different drives?is their any performance impact? |
| Answer |
| Yes you can expand but there will be performance impact. RAID groups can be only expanded with the same type of drives. |
|
| Question |
| Is their any way to make a single disk unbound from a raid group? |
| Answer |
| No, You cannot unbound a single disk from the RAID group. You need to destroy the complete RAID group. If you have any further queries, please get in touch with me at shivam.chawla@wipro.com |
|
| Question |
| How to verify whether SPS is faulty or sense cable is removed ? In case, when cabling status is unknown and SPS is failing at weekly battery check. |
| Answer |
|
| You can triage the SPCollects and in the SPA_cfg_info.txt and SPB_cfg_info.txt file, you can check that if SPS is present in the storage or not. If it showing as present in that file, then it means the cable is removed or faulted and you need to replace only sense calbe. If SPS is showing status as removed then you need to reseat the SPS and check the status and if it not comping up, you need to replace the SPS along with sense cable as we cannot judge in the case of SPS fault whether the sense cable is working fine or not. We can try replacing sense cable if SPS cabling status is unknown even after replacing the SPS. |
