Chinese
English
Problem description
Customers deployed osn3500 equipment, SSN3EAS2 single board and peg16 single board repeatedly flashed lsr_will_die alarm, the duration of about 20 seconds, after modifying the paranoid current threshold value, the network still has some optical ports reported this alarm, causing the customer to attach great importance to the customer, and asked to give the root cause of the analysis and a thorough solution.
Host version: v200r011c00spc200
Data board: ssn1peg16, ssn1pex1
Host version: v200r011c00spc200
Data board: ssn1peg16, ssn1pex1
Warning Messagelsr_will_die
ProcessingTemporary solution: Block the lsr_will_die alarm on the ge and 10ge ports on the network management, set the alarm threshold to 900, turn on the network management lsr_bcm_alm alarm monitoring, and determine whether the optical module is abnormal by observing the lsr_bcm_alm alarm.
Complete solution: need to develop a patch or subsequent version to solve the floating point operation exception problem
Complete solution: need to develop a patch or subsequent version to solve the floating point operation exception problem
Root Causeler_will_die alarm is the alarm reported by the paranoia current exceeding the set threshold value, this alarm indicates that the laser life is about to end, the alarm does not mean that the optical module immediately fails, but can continue to be used for a period of time, during this period of time to be ready to replace the optical module.
Combined with the phenomenon of the current network and the first line of data collection and analysis, more than one site at the same time reported that the alarm, and the use of optical modules not more than 2 years (the life of the optical module is generally in the 3 ~ 5 years), as well as this alarm is flashing, so the preliminary inference that more than one optical module at the same time the possibility of bad is not large, so from the single board software and hardware from the two sides to locate the problem.
Hardware:
1, analyze the single-board manufacturing information, found that some of the optical modules are not supported by the data veneer, and the same veneer using a lot of different vendors of optical modules, there is a certain degree of hidden danger, but is not the key to the problem, because the problem of the optical module in other sites are also working.
2, in the laboratory to build a mirror environment, reproduce the alarm.
3, the laboratory to complete the warm box test, to observe the relationship between the temperature and the alarm.
Software:
1, software R & D troubleshooting code, lsr_will_die alarm and lsr_bcm_alm alarm will theoretically appear in pairs, although the current network will be lsr_bcm_alm blocked, but you can use navigator to view this alarm.
2, coordinate a line, in the present network will bias current alarm high threshold set to 1, after 25 seconds or so to report the alarm, found that the bias current high alarm and laser life is about to end alarm are pairs of appearance, but in the present network of the previous alarm in the log, and did not find that the two alarms are pairs of appearances, the only difference between the two alarms in the judgment of the more floating-point operation, so the suspicion of floating-point operation is very big.
3, through the laboratory simulation, found and analyze the conclusion is consistent. Therefore, it can be judged that lsr_will_die alarm is due to abnormal floating-point operation caused by the alarm false alarms, previously other bureaus have also appeared in the floating-point operation abnormality caused by the power_abnormal alarm abnormal reporting problems in the laboratory to hang a temporary version of the floating-point operation results print, also appeared in the floating-point operation result of the anomaly.
At this point, it can be determined that the lsr_will_die alarm is a false alarm caused by floating-point operation anomalies.
Combined with the phenomenon of the current network and the first line of data collection and analysis, more than one site at the same time reported that the alarm, and the use of optical modules not more than 2 years (the life of the optical module is generally in the 3 ~ 5 years), as well as this alarm is flashing, so the preliminary inference that more than one optical module at the same time the possibility of bad is not large, so from the single board software and hardware from the two sides to locate the problem.
Hardware:
1, analyze the single-board manufacturing information, found that some of the optical modules are not supported by the data veneer, and the same veneer using a lot of different vendors of optical modules, there is a certain degree of hidden danger, but is not the key to the problem, because the problem of the optical module in other sites are also working.
2, in the laboratory to build a mirror environment, reproduce the alarm.
3, the laboratory to complete the warm box test, to observe the relationship between the temperature and the alarm.
Software:
1, software R & D troubleshooting code, lsr_will_die alarm and lsr_bcm_alm alarm will theoretically appear in pairs, although the current network will be lsr_bcm_alm blocked, but you can use navigator to view this alarm.
2, coordinate a line, in the present network will bias current alarm high threshold set to 1, after 25 seconds or so to report the alarm, found that the bias current high alarm and laser life is about to end alarm are pairs of appearance, but in the present network of the previous alarm in the log, and did not find that the two alarms are pairs of appearances, the only difference between the two alarms in the judgment of the more floating-point operation, so the suspicion of floating-point operation is very big.
3, through the laboratory simulation, found and analyze the conclusion is consistent. Therefore, it can be judged that lsr_will_die alarm is due to abnormal floating-point operation caused by the alarm false alarms, previously other bureaus have also appeared in the floating-point operation abnormality caused by the power_abnormal alarm abnormal reporting problems in the laboratory to hang a temporary version of the floating-point operation results print, also appeared in the floating-point operation result of the anomaly.
At this point, it can be determined that the lsr_will_die alarm is a false alarm caused by floating-point operation anomalies.
Recommendations and SummaryThe data veneer optical port lsr_will_die alarm is designed with reference to the ptn product, and the traditional data veneer of mstp product does not have this alarm, so we should pay attention to the applicability of the alarm and the sensitivity of the alarm when designing the alarm. In particular, some sensitive overseas customers see this alarm is very important, and asked to give the root cause, giving us unnecessary trouble. It is recommended to modify the alarm name or block the alarm directly.
In addition, for the alarm threshold setting, there must be a unified and effective standard, it is best not to easily modify the threshold value, otherwise it is easy to cause customer resentment and question.
In addition, for the alarm threshold setting, there must be a unified and effective standard, it is best not to easily modify the threshold value, otherwise it is easy to cause customer resentment and question.
END
Selling the OSN3500 SSN3EAS2_Product Quotation_Sales Manufacturer_Product Characteristics_Product Description_Huawei SDH Transmission Equipment Sales
Supply OSN3500 SSN3EAS2_Troubleshooting_Installation and Tune-up_Technical Specifications_Technical Parameters_Huawei SDH Transmission Equipment Sales
