ATTO Technology 4500R Uživatelský manuál Strana 1

- 1 - Thecus i8500 iSCSI RAID SYSTEM User Manual

- 10 - execute the data transaction. Each CV can have different cache memory size by user’s modification/setting. LUN (Logical Unit Number) is a uniqu

- 11 - Storage protocol, such as iSCSI, has “two ends” in the connection. These ends are initiator and target. In iSCSI, we call them iSCSI initiator

- 12 - 2. Linux iSCSI initiator is also available. For different kernels, there are different iSCSI drivers. Please check Appendix A for iSCSI initia

- 13 - Or login with read-only account which only allows reading the configuration and cannot change setting. Login name: user Default password: 1234

- 14 - When there is WARNING or ERROR occurred (LCM default filter), the LCM shows the event log to give users more detail from front panel. The foll

- 15 - [Quick Install] RAID 0 RAID 1 RAID 3 RAID 5 RAID 6 RAID 0+1 xxx GB [Apply The Config] [cYes Nod] [Local] RAID 0 RAID 1 RAID 3 RAID 5

- 16 - 2.4.2 System buzzer The system buzzer features are listed below: 1. The system buzzer alarms 1 second when system boots up successfully. 2.

- 17 - Chapter 3 Web GUI guideline 3.1 Web GUI hierarchy The below table is the hierarchy of web GUI. Quick installation ÆStep 1 / Step 2 / Confi

- 18 - UpgradeÆBrowse the firmware to upgrade / Export configuration Reset to defaultÆSure to reset to factory default? Import andexportÆImport/Expo

- 19 - Figure 3.2.2 1. RAID light: Green means RAID works well. Red represents RAID failure. 2. Temperature light: Green means normal temperatu

- 2 - Preface About this manual This manual is the introduction of i8500 and it aims to help users know the operations of the disk array system eas

- 20 - Step 2: Confirm page. Click “ ” if all setups are correct. Then a VD will be created. Done. You can start to use the system now. Figure 3.3

- 21 - Figure 3.4.1.1 Check “Change date and time” to set up the current date, time, and time zone before using or synchronize time from NTP (Netwo

- 22 - 3.4.3 Login setting “Login setting” can set single admin, auto logout time and Admin/User password. The single admin can prevent multiple user

- 23 - Figure 3.4.4.1 3.4.5 Notification setting “Notification setting” can set up SNMP trap for alerting via SNMP, pop-up message via Windows me

- 24 - Figure 3.4.5.1 “SNMP” allows up to 3 SNMP trap addresses. Default community setting is “public”. User can choose the event log levels and d

- 25 - Using “System log server”, user can choose the facility and the event log level. The default port of syslog is 514. The default setting enable

- 26 - 3.5.2 NIC “NIC” can change IP addresses of iSCSI data ports. Figure 3.5.2.1 (Figure 3.5.2.1: there are 2 iSCSI data ports.) IP settings: Use

- 27 - 3.5.3 Node “Node” can view the target name for iSCSI initiator. The node name of i8500 exists by default and can not be changed. Figu

- 28 - 4. Go to “/ iSCSI configuration / CHAP account” page to create CHAP account. Please refer to next section for more detail. 5. In “Authenticate”

- 29 - 2. Enter “User”, “Secret”, and “Confirm” secret again. Figure 3.5.5.3 3. Click “ ”. Figure 3.5.5.1 (Figure 3.5.5.4: create a CHAP account

- 3 - Table of Contents 1.1 Features... 5 1.2 Terminology ...

- 30 - RAID levels, it will use all available HDDs for RAID level depends on which user chooses. When system has different sizes of HDDs, e.g., 8*200G

- 31 - Figure 3.6.1.2 Step 3: Decide VD size. User can enter a number less or equal to the default number. Then click “ ”. Figure 3.6.1.3 Step 4:

- 32 - 3.6.2 Physical disk “Physical disk” can view the status of hard drives in the system. The followings are operational tips: 1. Mouse moves to

- 33 - Figure 3.6.2.3 (Figure 3.6.2.3: Physical disks of slot 1,2,3 are created for a RG named “RG-R5”. Slot 4 is set as dedicated spare disk of RG

- 34 - “Good” Æ the hard drive is good. “Failed” Æ the hard drive is failed. “Error Alert” Æ S.M.A.R.T. error alert. “Read Errors” Æ the hard drive h

- 35 - Set Dedicated spares Set hard drive to dedicated spare of selected RGs. Set property Change the status of write cache and standby. Write cache

- 36 - Step 2: Confirm page. Click “ ” if all setups are correct. Figure 3.6.3.2 (Figure 3.6.3.2: There is a RAID 0 with 4 physical disks, named “

- 37 - “Migrate” Æ the RAID group is being migrated. “Scrub” Æ the RAID group is being scrubbed. Health The health of RAID group. “Good” Æ the RAID gr

- 38 - information 3.6.4 Virtual disk “Virtual disk” can view the status of each Virtual disk. The following is an example to create a VD. Step 1

- 39 - Figure 3.6.4.2 (Figure 3.6.4.2: Create a VD named “VD-01”, related to “RG-R0”, size is 30GB. The other VD is named “VD-02”, initializing to 1

- 4 - 3.8.1 System information...54 3.8.2 Upgrade...

- 40 - Bg rate Background task priority. “4 / 3 / 2 / 1 / 0” Æ Default value is 4. The higher number the background priority of a VD has, the more ba

- 41 - #Snapshot Number of snapshot(s) that Virtual disk is taken. RG name The Virtual disk is related to the RG name Readahead This feature makes dat

- 42 - Cleanup snapshot Clean all snapshot VD related to the Virtual disk and release snapshot space. Take snapshot Take a snapshot on the Virtual dis

- 43 - Step 3: Take a snapshot. In “/ Volume configuration / Snapshot”, click “ ”. It will link to next page. Enter a snapshot name. Figure 3.6.5.

- 44 - • Snapshot column description: No. Number of this snapshot VD. The button next to the snapshot VD No. shows the functions which can be execut

- 45 - User can attach LUN by clicking the “ ”. “Host” must enter an iSCSI node name for access control, or fill-in wildcard “*”, which means every h

- 46 - Step 1: Create RG (RAID group). To create the RAID group, please follow the procedures: Figure 3.6.7.1 1. Select “/ Volume configuration /

- 47 - Figure 3.6.7.3 5. Select “/ Volume configuration / Virtual disk”. 6. Click “ ”. 7. Input a VD name, choose a RG Name and enter a size of VD;

- 48 - 2. Input “Host” name, which is a FC node name for access control, or fill-in wildcard “*”, which means every host can access to this volume. C

- 49 - In “/ Volume configuration / Logical unit”, 1. Mouse moves to the gray button next to the LUN; click “Detach”. There will pop up a confirma

- 5 - Chapter 1 RAID introduction 1.1 Features i8500 controller features: • Front-end 2 GbE NIC ports. • iSCSI jumbo frame support. • RAID 6, 60

- 50 - 1. Select “/ Volume configuration / Physical disk”. 2. Mouse moves to the gray button next to the PD slot; click “Set Free disk”. Step 10: Do

- 51 - 3.7.2 Hardware monitor “Hardware monitor” can view the information of current voltage and temperature. Figure 3.7.2.1 If “Auto shutdown” ha

- 52 - 3. If the high temperature situation doesn’t last for 3 minutes, controller will not do auto shutdown. 3.7.3 Hard drive S.M.A.R.T. support

- 53 - 3.7.4 UPS “UPS” can set up UPS (Uninterruptible Power Supply). Figure 3.7.4.1 Currently, the system only supports and communicates with sma

- 54 - “UPS shutdown in progress” “Batteries failed. Please change them NOW!” Battery Level (%) Current percentage of battery level. 3.8 System maint

- 55 - Figure 3.8.2.1 Figure 3.8.2.2 When upgrading, there is a progress bar running. After finished upgrading, the system must reboot manually t

- 56 - 3.8.4 Import and export “Import and export” allows user to save system configuration values: export, and apply all configuration: import. For

- 57 - Figure 3.8.5.1 The event log is displayed in reverse order which means the latest event log is on the first page. The event logs are actua

- 58 - Chapter 4 Advanced operation 4.1 Rebuild If one physical disk of the RG which is set as protected RAID level (e.g.: RAID 3, RAID 5, or RAID 6

- 59 - Sometimes, rebuild is called recover; they are the same meaning. The following table is the relationship between RAID levels and rebuild.

- 6 - 1.2 Terminology The document uses the following terms: RAID RAID is the abbreviation of “Redundant Array of Independent Disks”. There ar

- 60 - arrays. JBOD The abbreviation of “Just a Bunch Of Disks”. No data protection. RG fails if any hard drive failures or unplugs. 4.2 RG migra

- 61 - Figure 4.2.2 (Figure 4.2.2: A RAID 0 with 4 physical disks migrates to RAID 5 with 5 physical disks.) Figure 4.2.3 (Figure 4.2.3: A RAID 0

- 62 - 4.3 VD Extension To extend VD size, please follow the procedures. 1. Select “/ Volume configuration / Virtual disk”. 2. Mouse moves to the g

- 63 - capture. The certain location, named as “Snap VD”, is essentially a new VD.which can be attached to a LUN provisioned to a host as a disk like

- 64 - Figure 4.4.1.1 (Figure 4.4.1.1: This is Snap VD, but it is not exported.) 7. Mouse moves to the gray button next to the Snapshot VD number;

- 65 - 2. For performance and future rollback, system saves snapshot with names in sequences. For example, three snapshots has been taken and named “

- 66 - 3. Done. It will take snapshots automatically. Figure 4.4.2.1 (Figure 4.4.2.1: It will take snapshots every month, and keep the last 32 snap

- 67 - 2. When making rollback, the original VD cannot be accessed for a while. At the same time, the system connects to original VD and snaps VD, an

- 68 - use multiple physical paths to create logical "paths" between the server and the storage device. In the case which one or more of the

- 69 - Appendix A. Certification list • RAM i8500 RAM Spec: 184pins, DDR333(PC2700), Reg.(register) or UB(Unbufferred), ECC or Non-ECC, from 64MB t

- 7 - RAID width=1; RAID copy=4; RAID row=3. WT Write-Through cache-write policy. A caching technique in which the completion of a write request is no

- 70 - iSCSI initiator iscsi-initiator-utils-4.0.3.0-4 in kernel 2.6.9 3. For Red Hat Enterprise Linux 5 (Kernel 2.6), use the build-in iSCSI initiat

- 71 - Hitachi Deskstar 7K80, HDS728040PLA320, 40GB, 7200RPM, SATA II, 2M Hitachi Deskstar T7K500, HDT725032VLA360, 320GB, 7200RPM, SATA II, 16M Hit

- 72 - INFO SCSI info Received SCSI Bus Reset event at the SCSI Bus <number>. • EMS events Level Type Description INFO Power installed P

- 73 - INFO RG created RG <name> has been created. INFO RG creation failed Failed to create RG <name>. INFO RG deleted RG <name&

- 74 - write error occurred <address>-<address> of VD <name>. INFO Dedicated spare configured PD <slot> has been configured to

- 75 - backuped data finishes. INFO BBM detected Battery backup module is detected. INFO BBM is good Battery backup module is good. INFO BBM is char

- 76 - C. Known issues 1. Microsoft MPIO driver is not supported on Windows XP and Windows 2000 Professional. Workaround solution: Using Windows S

- 77 - D. Microsoft iSCSI Initiator Here is the step by step to setup Microsoft iSCSI Initiator. Please visit Microsoft website for latest iSCSI init

- 78 - Figure D.4 7. Click “Advance…” if CHAP information is needed. 8. Click “OK”. The status would be “Connected”. 9. Done, it can connect to an

- 79 - E. Installation steps for large volume (TB) Introduction: i8500 capable of supporting large volumes (>2TB). When connecting controllers to

- 8 - WWN World Wide Name. HBA Host Bus Adapter. SAF-TE SCSI Accessed Fault-Tolerant Enclosures. SES SCSI Enclosure Services. NIC Network Interface Ca

- 80 - 1. Follow the installation guild provided by HBA vendor, install HBA driver properly. For iSCSI models, please install the latest Microsoft iS

- 81 - http://www.microsoft.com/whdc/device/storage/GPT_FAQ.mspx Figure E.4 4. Format the disk. Figure E.5 5. Done. Figure E.6 6. The new d

- 82 - Figure E.7 7. Wrong setting result: OS can not format disk sector after 2048GB(2TB). Figure E.8 F. MPIO and MC/S setup instructio

- 83 - Figure F.1 The MPIO setup instructions are in the following: 1. Create RG/VD, and then attach LUN. 2. Add the first “Target Portal” on Mi

- 84 - 11. Click “Device” tab, then click “Advanced”. 12. Click “MPIO” tab, select “Load Balance Policy” to “Round Robin”. 13. Click “Apply”. 14. Run

- 85 - z INFO: Abnormal shutdown detected, start flushing battery-backed data (xxx KB). z INFO: Abnormal shutdown detected, flushing battery-backed

- 86 - H M/B Diagram

- 9 - RAID 0+1 Mirroring of the member RAID 0 volumes. RAID 0+1 needs at least four hard drives. RAID 10 Striping over the member RAID 1 volumes. RAID