UltraStor RS16 SCSI Ultra 320 to SATA IIa RAID Controller User Manual Version 3.0 (MAY.
Preface About this manual This manual is designed for UltraStor RS16 series controller user to operate the disk array system as easy as possible. Information contained in this manual has been checked for accuracy, but no product warranty is given with this content due to each operation environment is different. Information and specification are subject to change without further notice. Copyright@2006, Enhance Technology, Inc. All rights reserved. Thank you for using ENHANCE Technology, Inc.
Table of Contents Chapter 1 1.1 1.2 1.3 Features........................................................................... 5 Terminology ..................................................................... 6 RAID levels ...................................................................... 8 Chapter 2 2.1 2.2 2.3 Getting started ............................................... 9 Before starting.................................................................. 9 Storage introduction....................
Chapter 4 4.1 4.2 4.3 4.4 Rebuild........................................................................... 47 VG migration and expansion.......................................... 48 UDV Extension............................................................... 50 Snapshot/Rollback (optional) ......................................... 51 4.4.1 4.4.2 4.4.3 4.5 Advanced operation .................................... 47 Create snapshot volume ..................................................................
Chapter 1 RAID introduction 1.1 Features UltraStor RS16 controller is a high-performance RAID controller. • Backplane solution o UltraStor RS16: U320 SCSI (x2) -to- SATA II (x16) RAID controller. UltraStor RS16 controller has the features listed below: • • • • • • • • • • • RAID 6 ready. Snapshot-on-the-box / rollback (optional). SATA II support with SATA I backward compatible. N-way mirror. On-line volume expansion and RAID level migration. Global/dedicated cache configurable by volume. S.M.A.R.T.
UltraStor controller is the most cost-effective disk array controller with completely integrated high-performance and data-protection capabilities that meet or exceed the highest industry standards, the best data solution for small to medium size business user. Caution UltraStor series controller does not design SCSI terminator on system. It must add external SCSI terminators. Caution Snapshot/rollback features need 512MB RAM or more. Please refer to a certification list for RAM in Appendix A. 1.
GUI Graphic User Interface. RAID width, RAID copy, RAID row (RAID cell in one row) RAID width, copy and row are used to describe one VG. E.g.: 1. One 4-disk RAID 0 volume: RAID width= 4; RAID copy=1; RAID row=1. 2. One 3-way mirroring volume: RAID width=1; RAID copy=3; RAID row=1. 3. One RAID 10 volume over 3 4-disk RAID 1 volume: RAID width=1; RAID copy=4; RAID row=3. WT Write-Through cache write policy. Each data is synchronized in both data cache and the accessed physical disks.
SAF-TE SCSI Accessed Fault-Tolerant Enclosures. 1.3 RAID levels Describe RAID levels as follows: RAID 0 Disk striping. RAID 0 needs at least one hard drive. RAID 1 Disk mirroring over two disks. RAID 1 needs at least two hard drives. N-way mirror Extension to RAID 1 level. It has N copies of the disk. RAID 3 Striping with parity on the dedicated disk. RAID 3 needs at least three hard drives. RAID 5 Striping with interspersed parity over the member disks. RAID 3 needs at least three hard drives.
Chapter 2 Getting started 2.1 Before starting Before starting, prepare as follows. Review the “Certification list” in Appendix A to confirm that the hardware is fully supported. Read the latest release notes before upgrading. Release notes will accompany with firmware. A server with a SCSI HBA. SCSI cables and terminators. CAT 5, CAT 5e, or CAT 6 network cables for management port. Prepare storage system configuration plan. Management port network information.
2.3 Management methods There are three management methods to manage S Series controllers, describe on the following: 2.3.1 Web GUI S Series controllers support graphic user interface to manage the system. Be sure to connect LAN cable. The default IP is 192.168.0.1; so open the browser and type: http://192.168.0.200 Click any function at the first time; it will pop up a dialog to authenticate. Login name: admin Default password: 1234 2.3.
Tips It does not support telnet to connect for remote control, only ssh. Using ssh, the IP address has to be setup and the password is required for login. 2.4 Enclosure 2.4.1 LCM There are four buttons to control LCM (LCD Control Module), including: c (up), d (down), ESC (Escape), and ENT (Enter).
The following is LCM menu hierarchy. [Alarm Mute] [Reset/Shutdown] [Quick Install] ENHANCE Technology cd [View IP Setting] [cYes Nod] [Reset] [Shutdown] RAID 0 (RAID 1/RAID 3/ RAID 5/RAID 6) xxxxxx MB [IP Config] [Static IP] [IP Address] [192.168.000.001] [IP Subnet Mask] [255.255.255.0] [IP Gateway] [192.168.000.
2.4.1 System buzzer The system buzzer features are describing on the following: 1. 2. The system buzzer will alarm 3 seconds when system boots up successfully. The system buzzer will alarm continuously when there are error level events happened in the system. The alarm will be stopped after pressing mute.
Chapter 3 Web GUI guideline 3.
3.2 Login UltraStor series controller supports graphic user interface to manage the system. Be sure to connect LAN cable. The default IP is 192.168.0.200; so open the browser and type: http://192.168.0.200 Click any function at the first time; it will pop up a dialog to authenticate. Login name: admin Default password: 1234 After login, the selections listed on the left can be operated. Figure 3.2.1 There are three indicators at the top-right corner. 1. Voltage light: Green is normal.
3.3 Quick install It is easy to use “Quick install” function to create a volume. Depend on how many physical disks or how many residual spaces on created VGs are free, the system will calculate maximum spaces on RAID levels 0/1/3/5/6. “Quick install” function will occupy all residual VG space for one UDV, and it has no space for snapshot. If snapshot function is needed, please create volumes by manual, and refer to section 4.4 for more detail.
Figure 3.3.2 (Figure 3.3.2: A RAID 0 user data volume with the UDV name “QUICK20680”, named by the system itself, with the total available volume size 608256MB.) 3.4 System configuration “System config” selection is for the setup of “System name”, “IP address”, “Password”, “Date”, “Mail”, “SNMP” and view “Event log”.
Figure 3.4.1 3.4.1 System name Select “System name” to change system name. Default system name composed by model name and serial number of this system, e.g.: S120-000001. Figure 3.4.1.1 3.4.
Select “IP address” to change IP address for remote administration usage. There are 2 selections, DHCP (Get IP address from DHCP server) or set static IP. Figure 3.4.2.1 3.4.3 Password Select “Password” is for changing administrator password. Figure 3.4.3.1 3.4.4 Date Select “Date” to set up the current date and time before using. Figure 3.4.4.
3.4.5 Mail Select “Mail” to enter at most 3 mail addresses for receiving the event notification. Some mail servers would check “Mail-from address” and need authentication for anti-spam. Please fill the necessary fields and select “Send test mail” to check whether the email works fine. Figure 3.4.5.1 3.4.6 SNMP Select “SNMP” to set up SNMP trap for alert via SNMP. It allows up to 3 SNMP trap addresses can be set for receiving SNMP trap. Default community setting is “public”. Figure 3.4.6.
3.4.7 Event log Select “Event log” to view the event messages. Press “Clear” button will clear event log. Press “Mute” button will stop alarm if system alerts. Figure 3.4.7.1 3.5 Volume configuration “Volume config” selection is for the setup of volume configurations including “Physical disk”, “Volume group”, “User data volume”, “Cache volume”, and “Logical unit” functions.
Figure 3.5.1 3.5.1 Volume relationship diagram LUN 1 UDV 1 LUN 2 LUN 3 UDV 2 Snap UDV + + + VG PD 1 PD 2 Dedicated CV Global CV PD 3 DS RAM The above diagram describes the relationship of RAID components. One VG (Volume Group) consists of a set of UDVs (User Data Volume) and owns one RAID level attribute. Each VG could be divided into different UDVs. The UDVs from one VG share the same RAID level, but may own the different volume capacity.
4. The operations of the other web pages (e.g.: volume config of VG, UDV, CV, LUN pages) are similar. Figure 3.5.2.1 (Figure 3.5.2.1: Physical disks of slot 1, 2 have been created for a VG named “VG-R0”. Physical disks of slot 3, 4, 5 have been created for a VG named “VG-R5”. Slot 6 has been set as global spare disk. Slot 7 has been set as dedicated spare disk of VG named “VGR5”. Slot 8 is a free disk.) • PD column description: Slot The position of hard drives.
“DEFECT” Æ the hard drive has the bad blocks. “FAIL” Æ the hard drive cannot work in the respective volume. Status 1 “RD” Æ RAID Disk. This hard drive has been set to RAID. “FR” Æ FRee disk. This hard drive is free for use. “DS” Æ Dedicated Spare. This hard drive has been set to the dedicated spare of the VG. “GS” Æ Global Spare. This hard drive has been set to a global spare of all VGs. “RS” Æ ReServe. The hard drive contains the VG information but cannot be used.
Figure 3.5.3.1 (Figure 3.5.3.1: There is a RAID 0 with 2 physical disks, named “VG-R0”, total size is 152064MB, free size is 52096MB, related to 1 UDV. Another is a RAID 5 with 3 physical disks, named “VG-R5”.) No. Number of volume group. The blue square button next to the No. is “More Information” indication. It shows the details of the volume group. Name Volume group name. The blue square button next to the Name is “Rename” function. Total(MB) Total capacity of this volume group.
• Status 3 “M” Æ Migration. This volume group is doing migration. RAID The RAID level, which this volume group is using. The blue square button next to the RAID level is “Migrate” function. Click “Migrate” can add disk(s) to do expansion or change the RAID level of the Volume group.
3.5.4 User data volume Enter “User data volume” function to view the status of each user data volume. Figure 3.5.4.1 (Figure 3.5.4.1: Create a UDV named “UDV-R0”, related to “VG-R0”, size is 99968MB, status is online, write back, high priority, related to 1 LUN, with cache volume 80MB, no snapshot space. The other UDV is named “UDV-R5”, initializing to 53%) • UDV column description: No. Number of this user data volume. The blue square button in below to the UDV No. is “More Information” indication.
Status 1 “WT” Æ Write Through. “WB” Æ Write Back. The blue square button in below to the status1 is “Set read/write mode” function. Status 2 “HI” Æ HIgh priority. “MD” Æ MiD priority. “LO” Æ LOw priority. The blue square button in below to the status2 is “Set Priority” function. Status 3 “I” Æ user data volume is doing initializing. “R” Æ user data volume is doing rebuilding. Status 4 “M” Æ user data volume is doing migration. R% Ratio of initializing or rebuilding.
3.5.5 CREATE Create a user data volume function. DELETE Delete this user data volume function. Cache volume Enter “Cache volume” function to view the status of cache volume. The global cache volume is a default cache volume, which has been created after power on automatically, and cannot be deleted. The size of global cache is base on the RAM size. It will be total memory size minus that system uses. Figure 3.5.5.1 • CV column description: No. Number of this Cache volume.
• 3.5.6 CV operations description: CREATE Create a cache volume function. DELETE Delete this cache volume function. Logical unit number Enter “Logical unit” function to view the status of attached logical unit number of each UDV. Figure 3.5.6.1 • LUN operations description: ATTACH Attach a logical unit number to a user data volume. DETACH Detach a logical unit number from a user data volume. Caution Notify that which bus the SCSI cable is connected; it must match the bus ID which is attached.
3.5.7 Examples Take 2 examples to create volumes. Example 1 is to create two UDVs shared the same CV (global cache volume) and set a global spare disk. Example 2 is to create two UDVs. One shares global cache volume, the other uses dedicated cache volume. Set a dedicated spare disk. • Example 1 Example 1 is to create two UDVs in one VG, each UDV uses global cache volume. Global cache volume has been created after system boots up. So it doesn’t do anything about CV. Then set a global spare disk.
(Figure 3.5.7.2: Creating a RAID 5 with 4 physical disks, named “VG-R5”. The total size is 228096MB. Because of no related UDV there, free size still remains 228096MB.) Step 2: Create UDV (User Data Volume). To create a data user volume, please follow the procedures. Figure 3.5.7.3 1. Select “/ Volume config / User data volume”. 2. 3. Click “ ”. Input a UDV name, choose a VG Name and input a size to be used; decide the stripe high, block size, read/write mode and set priority, 4. 5. “.
(Figure 3.5.7.4: Create UDVs named “UDV-R5-1” and “UDV-R5-2”, related to “VG-R5”, the size of “UDV-R5-1” is 9984MB (it’s multiple of base stripe height, so the number may not be the same as the setting size.), the size of “UDV-R5-2” is 10944MB. The status of these UDVs are online, write back, high priority with cache volume 100MB. “UDV-R5-2” is initialing about 91%. There is no LUN attached.) Step 3: Attach LUN to UDV. There are 2 methods to attach LUN to UDV. 1.
Caution Be careful to avoid conflicts between SCSI ID at the same SCSI bus. Step 4: Set global spare disk. To set global spare disks, please follow the procedures. 1. 2. Select “/ Volume config / Physical disk”. Select the free disk(s) by clicking the checkbox of the row, then click 3. “ “ to set as global spares. There is a “GS” icon shown up at status 1 column. Figure 3.5.7.7 (Figure 3.5.7.7: Slot 5 has been set as global spare disk.) Step 5: Done. They can be used as SCSI disks.
Figure 3.5.7.8 1. Select UDVs by clicking the checkbox of the row, then click 2. 3. “ ”. There will pop up a confirm page. Choose “OK”. Done. Step 7: Delete UDV (User Data Volume). To delete the user data volume, please follow the procedures: 1. 2. Select “/ Volume config / User data volume”. Select UDVs by clicking the checkbox of the row. 3. 4. 5. Click “ “. There will pop up a confirm page. Choose “OK”. Done. The UDVs have been deleted.
Tips The action of deleting one VG will succeed only when all of the related UDV(s) are empty in this VG. Otherwise, it will have an error when deleting this VG. Step 9: Free global spare disk. To set global spare disks, please follow the procedures. 1. 2. Select “/ Volume config / Physical disk”. Select the global spare disk by clicking the checkbox of the row, then click “ “ to free disk. Step 10: Done, all volumes have been deleted. • Example 2 Example 2 is to create two UDVs in one VG.
Figure 3.5.7.9 1. 2. Select “/ Volume config / Cache volume”. If there is no free space for creating a new dedicated cache volume, cut down the global cache size first by clicking the blue square button “ ” in the size column. After resized, click “ volume page. ” to return to cache “ to enter the setup page. 3. Click “ 4. 5. “. Fill in the size and click “ Done. A new dedicated cache volume has been set. Tips The minimum size of global cache volume is 40MB.
2. 3. 4. Click “ ”. Input a UDV name, choose a VG Name, select Dedicated cache which is created at Step 1, and input a size to be used; decide the stripe height, block size, read/write mode and set priority, finally click “. “ Done. A UDV using dedicated cache has been created. Figure 3.5.7.11 (Figure 3.5.6.11: UDV named “UDV-R5-1” uses global cache volume 40MB, and “UDV-R52” uses dedicated cache volume 20MB. “UDV-R5-2” is initialing about 17%.) Figure 3.5.7.12 (Figure 3.5.6.
To set dedicated spare disks, please follow the procedures: 1. 2. Select “/ Volume config / Physical disk”. Select a VG from the list box, then select the free disk(s), click 3. “ ” to set as dedicated spare for the selected VG. There is a “DS” icon shown up at status 1 column. Figure 3.5.7.13 (Figure 3.5.7.13: Slot 5 has been set as dedicated spare disk of VG named “VG-R5”.) Step 6: Done. They can be used as SCSI disks. Delete UDVs, VG, please follow the steps. Step 7: Detach LUN from UDV.
2. Select the dedicated spare disk by clicking the checkbox of the row, then click “ “ to free disk. Step 11: Delete dedicated cache volume. To delete the cache volume, please follow the procedures: 1. 2. Select “/ Volume config / Cache volume”. Select a CV by clicking the checkbox of the row. 3. 4. 5. Click “ “. There will pop up a confirm page. Choose “OK”. Done. The CV has been deleted. Caution Global cache volume cannot be deleted. Step 12: Done, all volumes have been deleted. 3.
Figure 3.6.1 3.6.1 SAF-TE configuration SAF-TE represents SCSI Accessed Fault-Tolerant Enclosures, one of the enclosure management standards. Enter “SAF-TE config” function can enable or disable the management of SAF-TE from buses. Figure 3.6.1.1 (Figure 3.6.1.1: Enable SAF-TE in Bus 0, SCSI ID 15, and LUN 0.) The SAF-TE client software is available at the following web site: safte-monitor: http://oss.metaparadigm.com/safte-monitor/ SANtools: http://www.santools.
3.6.2 Voltage and Temperature Enter “Voltage & Temperature” function to view the information of current voltage and temperature. Figure 3.6.2.1 (for S50C/S100C) Figure 3.6.2.2 (for S60C/S120C) If “Auto shutdown” has been checked, the system will shutdown automatically when voltage or temperature is out of the normal range.
3.6.3 Hard drive S.M.A.R.T. function support S.M.A.R.T. (Self-Monitoring Analysis and Reporting Technology) is a diagnostic method for hard drives to give advanced warning of drive failures. Administrators wanted to know in advance if a hard drive was going to fail, because this gave them the opportunity to take steps to protect their data. S.M.A.R.T. measures many attributes of the hard drives over time and those hard drives can be decided if they are moving out of tolerance.
Figure 3.6.4.1 Currently, the system only support and communicate with smart-UPS function of APC (American Power Conversion Corp.) For UPS, Please take reference from http://www.apc.com/. First, interconnect via RS-232 cable between the system and UPS in order that the system can communicate with APC UPS. Then set up the shutdown values when the power is broken. UPS of other vendors can work fine, but they have no such function. UPS Type Select UPS Type.
“Batteries failed. Please change them NOW!” Battery Level Current percentage of battery level. (%) 3.7 System maintenance “Maintenance” function allows operation of the system functions including “Upgrade” to the latest firmware, “Info” to show the system version and “Shutdown” to either reboot or shutdown the system. Figure 3.7.1 3.7.1 Upgrade Enter “Upgrade” function to upgrade firmware. Please prepare new firmware file ” to select the file. named “xxxx.
3.7.2 Info Enter “Info” function will display firmware version. 3.7.3 Shutdown Enter “Shutdown” function; it will display “REBOOT” and “SHUTDOWN” buttons. Before power off, it’s better to press “SHUTDOWN” to flush the data from cache to physical disks. Figure 3.7.3.1 3.8 Logout For security reason, “Logout” function will allow logout while none is operating the system. Re-login the system by entering username and password.
Chapter 4 Advanced operation 4.1 Rebuild If one physical disk of the VG which sets to protected RAID level (e.g.: RAID 3 or RAID 5) is FAILED or has been plugged out, the VG becomes degrade mode, then the system will detect spare disk to rebuild the degrade VG to a complete one. It will detect dedicated spare as rebuild disk first, then global spare. In degrade mode, the status of VG will display “DG”.
RAID 3 Striping with parity on the dedicated disk. RAID 3 allows one hard drive fails or plugs out. RAID 5 Striping with interspersed parity over the member disks. RAID 5 allows one hard drive fails or plugs out. RAID 6 2-dimensional parity protection over the member disks. RAID 6 allows two hard drives fail or plug out. If it needs to rebuild two hard drives at the same time, it will rebuild the first one, then the other, by sequence. RAID 0+1 Mirroring of the member RAID 0 volumes.
4. Double check the setting of RAID level and RAID PD slot. If no problem, 5. click “ “. Finally a confirm page is shown with detail RAID info. If no problem, 6. “ to start migration. click “ Migration starts and it can be seen from the “status 3” of a VG with a running square and an “M”. In “/ Volume config / User data volume”, it will display a “M” in “Status 4” and complete percentage of migration in “R%”. Figure 4.2.1 Figure 4.2.2 (Figure 4.2.
Tips Executing migration/expansion, the total size of VG must be larger or equal to the original VG. It does not allow expanding the same RAID level with the same hard disks of original VG. Caution VG Migration cannot be executed during rebuild or UDV extension. 4.3 UDV Extension To extend UDV size, please follow the procedures. 1. 2. 3. 4. Select “/ Volume config / User data volume”. Decide which UDV to be extended, click the blue square button “ ” in the Size column next the number. Change the size.
Tips The size of UDV extension must be larger than original. Caution UDV extension cannot be executed during rebuild or migration. 4.4 Snapshot/Rollback (optional) Snapshot function will freeze the data at the moment while taking snapshot. When executing snapshot, it will become a new snap UDV, which can be attached a LUN then use it as a disk. Rollback function can return the whole data back to the time taking snapshot in order to avoid virus intrusion or files deletion by accident.
4.4.1 Create snapshot volume To take a snapshot of the data, please follow the procedures. 1. 2. 3. Select “/ Volume config / User data volume”. Choose a UDV to execute the snapshot by clicking the blue square button “ ” in the “Snapshot (MB)” column, it will direct to a setup page. Set up the size for executing snapshot. The size is suggested to be 4. 20% of UDV size or more, then click “ “. It will go back to the UDV page and the size will be shown in snapshot column.
3. 4. 5. 6. 4.4.2 created as name “snap1”(first), “snap2” and “snap3”(last). When deleting “snap2”, both of “snap1” and “snap2” will be deleted because “snap1” are related to “snap2”. For resource concern, the max number of snapshots is 32. If snapshot space is full, snap UDV will fail. Snap UDV cannot be migrated, when executing migration of related VG, snap UDV will fail. Snap UDV cannot be extended. Auto snapshot The snapshot copies can be taken manually or by schedule such as hourly and daily.
Tips Daily snapshot will be taken at every 00:00. Weekly snapshot will be taken on every Monday 00:00. 4.4.3 Rollback The data in snapshot UDV can rollback to original UDV. Please follow the procedures. 1. 2. 3. Select “/ Volume config / User data volume”. Take one or more snapshots. Please refer to section 4.4.1 for more detail. Click “ ” in “Snapshot (MB)” column to rollback the data, which is at the time of taking snapshot. Rollback function has some constraints as described in the following: 1. 2.
1. 2. Check the firmware of two systems first. It’s better that both have same firmware version or newer. Whole physical disks of related VG should be moved from system-1 to system-2. The configuration of both VG and UDV will be kept but LUN configuration will be cleared to avoid conflict with system-2. Appendix A. Certification list • RAM RAM Spec: 184pins, DDR333 (PC2700), Reg. (register) or UB (unbuffer), ECC or Non-ECC.
Vendor Model Hitachi Hitachi Hitachi Hitachi Hitachi Hitachi Hitachi Maxtor Samsung Seagate Seagate Seagate Seagate Seagate Seagate Seagate Seagate Western Digital Western Digital Western Digital Western Digital Deskstar 7K250, HDS722580VLSA80, 80GB, 7200RPM, SATA, 8M Deskstar 7K80, HDS728080PLA380, 80GB, 7200RPM, SATA-II, 8M Deskstar 7K160, 160GB, 7200RPM, SATA-II, 8M, NCQ Deskstar 7K 400, 400GB, 7200RPM, SATA-II, 8M Deskstar 7K500, HDS725050KLA360, 500G, 7200RPM, SATA-II, 16M Deskstar T7K500, 500G, 720
• HDD IO events Level Warning Warning Warning Warning • Disk error Disk error HDD failure Channel error Description Error: Disk read block error. Error: Disk writes block error. Error: Disk is failed. Error: Disk IO incomplete. SAF-TE events Level Info Warning Info Info • Type Type SAF-TE load conf. OK SAF-TE Load Conf. Failure SAF-TE is disabled SAF-TE is enabled Description Info: SAF-TE configuration has been loaded. Error: Failed to load SAF-TE configuration.
• System config events Level Info Info Error Warning • Sys Config. Defaults Restored Sys NVRAM OK Sys NVRAM IO Failed Sys NVRAM is full Description Info: Default system configurations restored. Info: The system NVRAM is active. Error: Can’t access the system NVRAM. Warning: The system NVRAM is full.
Warning UDV Degraded Info Warning UDV Init OK UDV_OP Stop Initialization UDV IO Fault Warning Info VG Failed UDV Failed Global CV Adjustment Failed Global Cache Global CV Creation Failed UDV Rename Info VG Rename Info Set VG Dedicated Spare Disks Set Global Disks Warning Warning Warning Info Error Info Info Info Info Info Info Info • UDV Read-Only WRBK Cache Policy WRTHRU Cache Policy High priority UDV Mid Priority UDV Low Priority UDV Warning: UDV is under degraded mode.
C. Known issues 1. In Microsoft Windows server 2003 or Windows XP, there is at least a LUN 0 in each SCSI ID; otherwise, it cannot be recognized. Windows 2000 server has not such constraint. Workaround solution: attach LUN from LUN 0. 2. In Windows server 2003 or Windows XP, Because the Linux driver of Adaptec SCSI Card 29320A-R, 39320A-R has some problems, it cannot negotiate with Ultra 320 speed.
