Azure Linux虚拟机备份文件恢复时出现超时错误

问题现象 #

验证Azure Linux VM备份功能时，下载最新恢复点的恢复脚本执行后出现以下异常：

1. 脚本成功完成兼容性、网络等前置检查，建立iSCSI连接后，在自动挂载磁盘步骤卡住，停留在：

Microsoft Azure VM Backup - File Recovery
______________________________________________
Please verify the license, terms and conditions for using this script as mentioned here: https://aka.ms/OpenSourceLicenseForFileRecovery
                                                                                        ------------------------------------------------
Checking for OS Compatibility: DONE
Checking for large disks: DONE
Checking for LVMs: DONE
Checking for network connectivity: DONE
Checking for required cipher suite: DONE
Please enter the password as shown on the portal to securely connect to the recovery point. : XXXXXX
Connecting to recovery point using ISCSI service...

Connection succeeded!

Please wait while we attach volumes of the recovery point to this machine...
SCSI

2. 查看dmesg日志确认磁盘已被系统识别（如sdb、sdc设备）：

[  168.267791] Loading iSCSI transport class v2.0-870.
[  197.798995] iscsi: registered transport (tcp)
[  198.059221] scsi host3: iSCSI Initiator over TCP/IP
[  198.232121] scsi 3:0:0:0: Direct-Access  MABILR I Disk           1.00 PQ: 0 ANSI: 5
[  198.295800] sd 3:0:0:0: Attached scsi generic sg2 type 0
[  198.336224] scsi host4: iSCSI Initiator over TCP/IP
[  198.386131] sd 3:0:0:0: [sdb] 62914560 512-byte logical blocks: (32.2 GB/30.0 GiB)
[  198.400660] sd 3:0:0:0: [sdb] Write Protect is off
[  198.400669] sd 3:0:0:0: [sdb] Mode Sense: 27 00 10 04
[  198.429963] sd 3:0:0:0: [sdb] Write cache: disabled, read cache: disabled, supports DPO and FUA
[  198.530547] scsi 4:0:0:0: Direct-Access  MABILR I Disk           1.00 PQ: 0 ANSI: 5
[  198.591583] sd 4:0:0:0: Attached scsi generic sg3 type 0
[  198.632254] scsi host5: iSCSI Initiator over TCP/IP
[  198.681411] sd 4:0:0:0: [sdc] 1073741824 512-byte logical blocks: (550 GB/512 GiB)
[  198.696483] sd 4:0:0:0: [sdc] Write Protect is off
[  198.696490] sd 4:0:0:0: [sdc] Mode Sense: 27 00 10 04
[  198.726301] sd 4:0:0:0: [sdc] Write cache: disabled, read cache: disabled, supports DPO and FUA

3. 手动挂载rootfs后连接很快中断，dmesg出现I/O超时错误：

[  378.572336] sd 3:0:0:0: [sdb] tag#67 timing out command, waited 180s
[  378.577504] sd 3:0:0:0: [sdb] tag#67 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=180s
[  378.577517] sd 3:0:0:0: [sdb] tag#67 Sense Key : Unit Attention [current]
[  378.577525] sd 3:0:0:0: [sdb] tag#67 Add. Sense: No additional sense information
[  378.577549] sd 3:0:0:0: [sdb] tag#67 CDB: Read(10) 28 00 00 00 00 00 00 00 08 00
[  378.577566] blk_update_request: I/O error, dev sdb, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[  378.584439] Buffer I/O error on dev sdb, logical block 0, async page read

4. 无法检测并挂载任何LVM卷

已尝试的排查操作 #

• 在不同地域的两台测试机上验证Debian 12、CentOS 7、Ubuntu 22.04系统，问题复现一致
• 调整iscsid.conf中的noop_out_timeout参数（遵循微软文档建议），无改善

请问有没有人遇到过类似问题？或者知道可能的根因？

内容的提问来源于stack exchange，提问作者Gerrit