硬件更换流程.doc

1、 18DA 22501 4008 ASCVQ A1430 HLR 硬件更换流程文档编号 8DA 22501 4008 ASCVQ 文档版本 V2.5适应的运营商 CMCC/UNICOM 安全性 风险高作 者 张园庆 联系方式产 品 A1430 HLR 适应版本 R2, R3文档版本历史版本 日期 描述 作者1.0 2006-07-29 Creation of the document 张园庆1.1 2006-07-29 Update 赵荣东1.2 2007-08-28 Modify P4 红色字体 张园庆1.3 2008-07-28 Update key procedure. Lin Chun

2、yi1.4 2009-05-10 Update details LIN Chunyi2.0 2009/5/26 update Hua youping2.1 2009/7/13 增加微动开关的文字说明 Wang peng2.2 2009/10/30 增加 lock 板卡后需检查的状态 Wang peng2.3 2009/11/23 增加 F300 更换流程 Wang peng2.4 2009/11/27 update Lin chunyi2.5 2009/12/1 Add checkpoint before pilot replacement Zhao rongdong2.6 2010-2-9

3、Update procedure for checking hardware clock and stopping DB clusterZhu meng21. Control station 替换准备工作： 1) 用 df 命令检查各目录硬盘空间占用率小于 85%2) 更换硬件前检查系统中是否有自动备份在运行，如果有，请停止自动备份的运行。3) system , platform, all_data, mysql 数据备份。4) emergency restore 窗口已经下载。5) 新的 pilot 备板 BIOS 参数已经重定向设置成通过 COM 接口连接到 IOLAN 输出，并且不从硬盘

4、启动。检查新板卡微动开关的设置和下图一致：注意：只有 SW4 的第二个微动需要设置成 on，其他所有的微动都是 off。注意：以下图片的微动开关方向和板卡上的微动开关方向有可能不一致，以板卡上的微动开关的说明为准，例如：微动开关在板卡上的分布如下：36) PCFS 通过 IOLAN 用 telnet 连接到需要更换的 PILOT 板7) 注意静电防护，比如带上静电手套等等更换过程1.1 lock和set equipment off standby pilot(1)在 active pilot 上执行 fw lsta 命令检查所有板卡运行状态，确认所有板卡的状态都是好的：roothlrsz5st

5、b:# fw lsta*FWLSTA *host_name oper_state syst_state susp_req version STATION_A enabled standby no r3.310.1303-01 STATION_B enabled active no r3.310.1303-01 STATION_C enabled active no r3.310.1303-01 STATION_D enabled active no r3.310.1303-01 5STATION_E enabled active no r3.310.1303-01 STATION_F enab

6、led active no r3.310.1303-01 STATION_G enabled active no r3.310.1303-01 STATION_H enabled active no r3.310.1303-01 STATION_I enabled active no r3.310.1303-01 STATION_J enabled active no r3.310.1303-01 STATION_K enabled active no r3.310.1303-01 STATION_L enabled active no r3.310.1303-01 STATION_M dis

7、abled not_installed noSTATION_N disabled not_installed noSTATION_O disabled not_installed noSTATION_P disabled not_installed noSTATION_Q disabled not_installed noSTATION_R disabled not_installed noSTATION_S disabled not_installed noSTATION_T disabled not_installed noSTATION_U disabled not_installed

8、noSTATION_V disabled not_installed noSTATION_W disabled not_installed noSTATION_X disabled not_installed noroothlrsz5stb:#在 pilot 上执行 fw restart x (x 为 standby pilot)重启 standby pilot执行 fw lsta 命令检查 standby pilot 启动完成，确认 active / standby pilot。在 active pilot 上进入buii，执行命令：get staPILOT/*(*);检查 standby

9、pilot 的如下状态参数为：operational_state = enabled,system_state = standby, 检查 active pilot 的如下状态参数为：operational_state = enabled,system_state = active, get logical_nodeNGHLR/A(*);get logical_nodeNGHLR/B(*);检查 standby pilot 的如下状态参数为：functional_name = “Pilot“,adm_state = unlocked,op_state = enabled,failed = no

10、,in_test = no,power_off = no,offline = no,dependency = no,degraded = no,not_installed = no,st_status = hot_standby, node_type = tomas_station,node_subtype = tomas_pilot_station,检查 active pilot 的如下状态参数为：adm_state = unlocked,op_state = enabled,failed = no,in_test = no,power_off = no,6offline = no,depe

11、ndency = no,degraded = no,not_installed = no,st_status = providing_service,node_type = tomas_station,node_subtype = tomas_pilot_station,(2)在 active pilot 上使用 fw lock X(X=A or B)命令锁定 standby pilot：roothlrsz5stb:# fw lock A (假设 A 为 standby pilot 的情况)- OK, connection active- Number of clients = 14/* An

12、swer to request #1 = */ACTION_RSP logical_node NGHLR/A(lock();roothlrsz5stb:#(3)standby pilot 板卡被锁定后检查以下状态是否正确：roothlrsz5stb:# fw lsta*FWLSTA *host_name oper_state syst_state susp_req version STATION_A disabled not_installed no r3.310.1303-01 STATION_B enabled active no r3.310.1303-01 STATION_C enab

13、led active no r3.310.1303-01 STATION_D enabled active no r3.310.1303-01 STATION_E enabled active no r3.310.1303-01 STATION_F enabled active no r3.310.1303-01 STATION_G enabled active no r3.310.1303-01 STATION_H enabled active no r3.310.1303-01 STATION_I enabled active no r3.310.1303-01 STATION_J ena

14、bled active no r3.310.1303-01 STATION_K enabled active no r3.310.1303-01 STATION_L enabled active no r3.310.1303-01 STATION_M disabled not_installed noSTATION_N disabled not_installed noSTATION_O disabled not_installed noSTATION_P disabled not_installed noSTATION_Q disabled not_installed noSTATION_R

15、 disabled not_installed noSTATION_S disabled not_installed noSTATION_T disabled not_installed noSTATION_U disabled not_installed noSTATION_V disabled not_installed noSTATION_W disabled not_installed noSTATION_X disabled not_installed no7roothlrsz5stb:#roothlrsz5stb:# buii- OK, connection active- Num

16、ber of clients = 14BUI get staPILOT/A(*); (假设 A 为 standby pilot 的情况)Answer to request #1 =GET_RSP sta PILOT/A(/* Attributes : */creation_date = 2009/08/28 10:54:32,adm_upd_date = 2009/08/28 10:54:32,tsap = 65,tsap_mirror = 66,hostname = “STATION_A“,profile_name = “_pilot_“,type_sta = pilot,syst_upd_

17、date = 2009/10/30 16:05:12,operational_state = disabled,system_state = not_installed,suspend_requested = no,version = “r3.310.1303-01“) ;The request number 1 has been sent to localhostBUI get logical_nodeNGHLR/A(*); (假设 A 为 standby pilot 的情况)Answer to request #2 =GET_RSP logical_node NGHLR/A(/* Attr

18、ibutes : */creation_date = 2006/06/09 21:52:59,adm_upd_date = 2007/08/10 02:27:12,syst_upd_date = 2009/10/30 16:05:13,functional_name = “Pilot“,adm_state = locked,op_state = disabled,failed = no,in_test = no,power_off = no,offline = yes,dependency = no,degraded = no,not_installed = no,st_status = co

19、ld_standby,node_type = tomas_station,node_subtype = tomas_pilot_station,node_name = “STATION_A“,tsap = 65,memory_size = 4,cpu_number = 2,8(4)给被锁定的 pilot 下电：BUI action logical_nodeNGHLR/A(set_equipment_off(); (假设 A 为 standby pilot 的情况)Answer to request #3 =ACTION_RSP logical_node NGHLR/A(set_equipmen

20、t_off();The request number 3 has been sent to localhostBUI(5)下电完成后检查以下红色标示部分状态是否正常：roothlrsz5stb:# fw lsta*FWLSTA *host_name oper_state syst_state susp_req version STATION_A disabled not_installed no r3.310.1303-01 STATION_B enabled active no r3.310.1303-01 STATION_C enabled active no r3.310.1303-01

21、 STATION_D enabled active no r3.310.1303-01 STATION_E enabled active no r3.310.1303-01 STATION_F enabled active no r3.310.1303-01 STATION_G enabled active no r3.310.1303-01 STATION_H enabled active no r3.310.1303-01 STATION_I enabled active no r3.310.1303-01 STATION_J enabled active no r3.310.1303-0

22、1 STATION_K enabled active no r3.310.1303-01 STATION_L enabled active no r3.310.1303-01 STATION_M disabled not_installed noSTATION_N disabled not_installed noSTATION_O disabled not_installed noSTATION_P disabled not_installed noSTATION_Q disabled not_installed noSTATION_R disabled not_installed noST

23、ATION_S disabled not_installed noSTATION_T disabled not_installed noSTATION_U disabled not_installed noSTATION_V disabled not_installed noSTATION_W disabled not_installed noSTATION_X disabled not_installed noroothlrsz5stb:# buii- OK, connection active- Number of clients = 14BUI get staPILOT/A(*); (假

24、设 A 为 standby pilot 的情况)9Answer to request #1 =GET_RSP sta PILOT/A(/* Attributes : */creation_date = 2009/08/28 10:54:32,adm_upd_date = 2009/08/28 10:54:32,tsap = 65,tsap_mirror = 66,hostname = “STATION_A“,profile_name = “_pilot_“,type_sta = pilot,syst_upd_date = 2009/10/30 16:05:12,operational_stat

25、e = disabled,system_state = not_installed,suspend_requested = no,version = “r3.310.1303-01“) ;The request number 1 has been sent to localhostBUI get logical_nodeNGHLR/A(*); (假设 A 为 standby pilot 的情况)Answer to request #2 =GET_RSP logical_node NGHLR/A(/* Attributes : */creation_date = 2006/06/09 21:52

26、:59,adm_upd_date = 2007/08/10 02:27:12,syst_upd_date = 2009/10/30 16:13:33,functional_name = “Pilot“,adm_state = locked,op_state = disabled,failed = no,in_test = no,power_off = no,offline = yes,dependency = no,degraded = no,not_installed = yes,st_status = cold_standby,node_type = tomas_station,node_

27、subtype = tomas_pilot_station,node_name = “STATION_A“,tsap = 65,memory_size = 4,cpu_number = 2,注意：standby pilot 被 lock 住后状态必须是 not_installed，如果 standby pilot 被 lock 住后状态仍然是 standby，请停止当前的所有操作，并立即联系相关区域的 TAC 负责人！101.2 替换硬件 拨出要替换的板卡：注意先松开板卡下面的扳手，等蓝灯常亮之后再拔出； 插入新的 PILOT 后，立即在 PCFS 上通过 IOLAN 用 telnet 连接到

28、需要更换的PILOT 板，按 F2 进入 BIOS。如果进入 BIOS 没有成功，应立即启动该PILOT 的 reset 按钮，并在 PCFS 上按 F2 进入 BIOS，记录该 PILOT 的两个MAC 地址，设置该板卡的 BIOS，BIOS 参数设置见附录 4.1，不要退出。 在新插入的 PILOT 板 BIOS 中检查并修改确认系统不从硬盘启动，在 Hard Drive 前加上“！”，并保存退出。注意：如果有 CF 卡，需要将 CF 卡启动放在硬盘启动的前面并去掉卡启动前面的”!”，如下图：Hard Drive 下面 TRANSCEND-(SM)表示 CF 卡启动1.3 修改备用侧的MA

29、C 地址登录到 active pilot： 在Linux 下， 在#后输入以下命令并回车/usr/nectar/bin/nse/nse_clientupdate station-name 说明： station-name is standby pilot; example: station_A. or .station_B 选择2 修改 ethernet address 并回车。 择对应需要修改的pilot 名称并回车。 按照提示替换两个MAC 地址。11 确认MAC地址是否正确。如果检查一致，输入字母 Y并回车。 输入q结束配置。1.4 改备用侧PILOT的BIOS执行restore硬件 r

30、eset 刚插入的板卡 按住PCFS的F2 ，进入BIOS 设置。 在BOOT 选项中将ETH1 和ETH2 移到 Hard drive 之前。 保存退出，备用侧PILOT然后开始自动RESTORE。 等到restore结束（大约 30分钟），会出现以下信息：Do you want to reboot 输入Y ，回车。（请马上按住F2，进入BIOS设置）1.5 修改备用侧PILOT的BIOS 按住终端电脑的F2，进入BIOS 设置。 检查BIOS里的时间设置，如果和active pilot 时间不一致，请修改过来 在BOOT 选项中将ETH1 和ETH2 移到 Hard drive 之后。在Hard Drive前去掉“ ！”， 注意：如果pilot板卡安装了CF 卡，将CF卡前面的 ”!”去掉，并放在硬盘启动的前面！，设置完毕之后，如下图：TRANSCEND-(SM)放在第一启动位 卡启动FUJITSU MHV2060AS-(PM放在第二启动位)硬盘启动网卡启动放在后面(IBA GE Slot 012.)