No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

Rebooted Host Failed to Be Connected to the Storage Array Owing To Modification of CHAP User Information

Publication Date:  2013-06-28 Views:  77 Downloads:  0
Issue Description
Product and version information:
  • Storage system: S5000T V100R001 and V100R002
  • Host system:
− Model: Huawei RH2285
− Operating system: Microsoft Windows Server 2008 Enterprise R2 Edition
  •  FC HBA: Intel (R) Ethernet Server Adapter X520-2, driver version 2.9.66.0
The storage array and host form an IP SAN network as shown in following figure.
CHAP users, after created on the storage array, are added to the host initiator. On both host and storage array, the CHAP certification is enabled and the iSCSI connection is built. Then, Names of CHAP users on the storage array are modified. After the host is rebooted, the iSCSI link is down. The ISM displays that the initiator is not connected, as shown in the following figure.
Alarm Information
None
Handling Process
Manually set CHAP information on the host and rebuilt the iSCSI connection.
Root Cause
After the host is rebooted, the system attempts to recover the iSCSI link between the host and storage array. The host retains the original CHAP user information. However, the storage array uses the modified CHAP user information. The information inconsistency leads to a failure of CHAP certification. As a result, the iSCSI link cannot be recovered.
Compared with EMC arrays, operations workable on EMC arrays cannot recover the iSCSI link.
Suggestions
After the CHAP certification is enabled on both the host and storage array, you are advised not to modify CHAP user information on the storage array. If you have to do so, see step 3.

Reference Information
1. After the host is rebooted, capture packets on the storage array. The following figure shows the CHAP information.

As shown in red boxes of the figure, the CHAP user name carried by the Login Command of the host is test, while the one observed on the storage array is root. This is why the CHAP certification fails.
The packet capturing information shows that the host fails to be connected to the storage array (initiator failed to automatically recover the connection after the host is rebooted). This is because the host carried CHAP certification information differs from that on the storage array. Specifically, CHAP users are replaced on the storage array. However, the host uses the original CHAP users for certification after rebooted.
2. The same problem happens to EMC arrays.
As shown in the following figure, the EMC array is connected to the Windows Server 2008 host. When the CHAP certification is enabled, two links are normal.

Replace CHAP users on the storage array and reboot the host. The host fails to be connected to the EMC array, as shown in the following figure.

3. Finally, we have the solution: If you modify CHAP users on the storage array, you need to delete content in the red box as shown in the following figure. Then, rebuild the connection.
In this way, the connection is automatically recovered after the host is rebooted. Inside the red box is the negotiation content between the host initiator and storage array for reserving the CHAP certification information. If you only modify CHAP user information on the storage array without updating the negotiation content, the host will use the original CHAP certification information after rebooted, causing the certification to fail.

END