Hello everyone.
I'm trying to simulate the eruption of Changbai Mountain(Baekdu Mountain) following the WRF-Chem 3.9.1.1 Emissions Guide .(https://ruc.noaa.gov/wrf/wrf-chem/Emission_guide.pdf). But It always failed when I runed the convert_emiss program because memory corruption
following is my process:
I have successfully generate these *.bin files by PRE-CHEM-SRC(v1.5) and wrfinput_d01 by WRF(v4.3.3). I have checked these file and logs of processing in order to assure I can see 'SUCCESS' in these logs.
Then I copy them(*.bin,wrfinput_d01) to the WRF4-3-3/chem and WRF4-3-3/run, and modified namelist.input for convert emiss following the WRF-Chem 3.9.1.1 Emissions Guide.
Finally I run convert_emiss.exe, the issue happened.
I tried fix this problem for many times, for example:
-runing convert_emiss by *.sh file, and add the number of processors.
-changing the name of files generated by PRE-CHEM-SRC(That's because when I runed the PRE-CHEM-SRC, the files for WRF is named by matrixfire-T-{years}-{month}-{days}-000000-g1-*.bin, I checked log of it, it shows, for example, volcanic SO2 and ASH emission, height,elevation and duration data for WRF: matrixfire-T-2024-08-06-000000-g1-volc.bin. So I change them to meet the file name of WRF-Chem 3.9.1.1 Emissions Guide)
-changing the namelist for WRF and generate the wrfinput for many times since I think maybe my wrfinput have something wrong.
-searching this problem on the internet, but I only got that the memory issue often occur when running convert_emiss.
The report of it is here, and I would upload my namelist for convert_emiss:
DYNAMICS OPTION: Eulerian Mass Coordinate
alloc_space_field: domain 1 , 18282880052 bytes allocated
open file wrfinput_d01
med_initialdata_input: calling input_input
Input data is acceptable to use: wrfinput_d01
CURRENT DATE = 2024-08-06_00:00:00
SIMULATION START DATE = 2024-08-06_00:00:00
Timing for processing wrfinput file (stream 0) for domain 1: 1.39594 elapsed seconds
Max map factor in domain 1 = 1.06. Scale the dt in the model accordingly.
D01: Time step = 60.0000000 (s)
D01: Grid Distance = 12.0000000 (km)
D01: Grid Distance Ratio dt/dx = 5.00000000 (s/km)
D01: Ratio Including Maximum Map Factor = 5.30389738 (s/km)
D01: NML defined reasonable_time_step_ratio = 6.00000000
INPUT LandUse = "MODIFIED_IGBP_MODIS_NOAH"
LANDUSE TYPE = "MODIFIED_IGBP_MODIS_NOAH" FOUND 41 CATEGORIES 2 SEASONS WATER CATEGORY = 17 SNOW CATEGORY = 15
*** Error in `./convert_emiss.exe': malloc(): memory corruption (fast): 0x000000000f0f5500 ***
======= Backtrace: =========
/lib64/libc.so.6(+0x7f5d4)[0x2b3ffa04f5d4]
/lib64/libc.so.6(+0x82d10)[0x2b3ffa052d10]
/lib64/libc.so.6(__libc_malloc+0x4c)[0x2b3ffa0551ac]
./convert_emiss.exe[0x36aac28]
./convert_emiss.exe[0x2896073]
./convert_emiss.exe[0x28baae8]
./convert_emiss.exe[0x253f809]
./convert_emiss.exe[0x24e5770]
./convert_emiss.exe[0x20e6423]
./convert_emiss.exe[0x4077aa]
./convert_emiss.exe[0x406f71]
/lib64/libc.so.6(__libc_start_main+0xf5)[0x2b3ff9ff23d5]
./convert_emiss.exe[0x406fa1]
======= Memory map: ========
……
2b3ffc7bc000-2b3ffc806000 r-xp 00000000 fd:00 134257705 /usr/lib64/libgssapi_krb5.so.2.2
2b3ffc806000-2b3ffca06000 ---p 0004a000 fd:00 134257705 /usr/lib64/libgssapi_krb5.so.2.2
2b3ffca06000-2b3ffca07000 r--p 0004a000 fd:00 134257705 /usr/lib64/libgssapi_krb5.so.2.2
2b3ffca07000-2b3ffca09000 rw-p 0004b000 fd:00 134257705 /usr/lib64/libgssapi_krb5.so.2.2
2b3ffca09000-2b3ffcae2000 r-xp 00000000 fd:00 134257715 /usr/lib64/libkrb5.so.3.3
Program received signal SIGABRT: Process abort signal.
Backtrace for this error:
#0 0x2B3FF9573697
#1 0x2B3FF9573CDE
#2 0x2B3FFA00627F
#3 0x2B3FFA006207
#4 0x2B3FFA0078F7
#5 0x2B3FFA048D26
#6 0x2B3FFA04F5D3
#7 0x2B3FFA052D0F
#8 0x2B3FFA0551AB
#9 0x36AAC27 in __module_ra_cam_support_MOD_oznini
#10 0x2896072 in __module_physics_init_MOD_ra_init
#11 0x28BAAE7 in __module_physics_init_MOD_phy_init
#12 0x253F808 in start_domain_em_
#13 0x24E576F in start_domain_
#14 0x20E6422 in med_initialdata_input_
#15 0x4077A9 in MAIN__ at convert_emiss.f90:?
Could anyone please share the methods of solving this problem? That will be really helpful for me. Thanks.
I'm trying to simulate the eruption of Changbai Mountain(Baekdu Mountain) following the WRF-Chem 3.9.1.1 Emissions Guide .(https://ruc.noaa.gov/wrf/wrf-chem/Emission_guide.pdf). But It always failed when I runed the convert_emiss program because memory corruption
following is my process:
I have successfully generate these *.bin files by PRE-CHEM-SRC(v1.5) and wrfinput_d01 by WRF(v4.3.3). I have checked these file and logs of processing in order to assure I can see 'SUCCESS' in these logs.
Then I copy them(*.bin,wrfinput_d01) to the WRF4-3-3/chem and WRF4-3-3/run, and modified namelist.input for convert emiss following the WRF-Chem 3.9.1.1 Emissions Guide.
Finally I run convert_emiss.exe, the issue happened.
I tried fix this problem for many times, for example:
-runing convert_emiss by *.sh file, and add the number of processors.
-changing the name of files generated by PRE-CHEM-SRC(That's because when I runed the PRE-CHEM-SRC, the files for WRF is named by matrixfire-T-{years}-{month}-{days}-000000-g1-*.bin, I checked log of it, it shows, for example, volcanic SO2 and ASH emission, height,elevation and duration data for WRF: matrixfire-T-2024-08-06-000000-g1-volc.bin. So I change them to meet the file name of WRF-Chem 3.9.1.1 Emissions Guide)
-changing the namelist for WRF and generate the wrfinput for many times since I think maybe my wrfinput have something wrong.
-searching this problem on the internet, but I only got that the memory issue often occur when running convert_emiss.
The report of it is here, and I would upload my namelist for convert_emiss:
DYNAMICS OPTION: Eulerian Mass Coordinate
alloc_space_field: domain 1 , 18282880052 bytes allocated
open file wrfinput_d01
med_initialdata_input: calling input_input
Input data is acceptable to use: wrfinput_d01
CURRENT DATE = 2024-08-06_00:00:00
SIMULATION START DATE = 2024-08-06_00:00:00
Timing for processing wrfinput file (stream 0) for domain 1: 1.39594 elapsed seconds
Max map factor in domain 1 = 1.06. Scale the dt in the model accordingly.
D01: Time step = 60.0000000 (s)
D01: Grid Distance = 12.0000000 (km)
D01: Grid Distance Ratio dt/dx = 5.00000000 (s/km)
D01: Ratio Including Maximum Map Factor = 5.30389738 (s/km)
D01: NML defined reasonable_time_step_ratio = 6.00000000
INPUT LandUse = "MODIFIED_IGBP_MODIS_NOAH"
LANDUSE TYPE = "MODIFIED_IGBP_MODIS_NOAH" FOUND 41 CATEGORIES 2 SEASONS WATER CATEGORY = 17 SNOW CATEGORY = 15
*** Error in `./convert_emiss.exe': malloc(): memory corruption (fast): 0x000000000f0f5500 ***
======= Backtrace: =========
/lib64/libc.so.6(+0x7f5d4)[0x2b3ffa04f5d4]
/lib64/libc.so.6(+0x82d10)[0x2b3ffa052d10]
/lib64/libc.so.6(__libc_malloc+0x4c)[0x2b3ffa0551ac]
./convert_emiss.exe[0x36aac28]
./convert_emiss.exe[0x2896073]
./convert_emiss.exe[0x28baae8]
./convert_emiss.exe[0x253f809]
./convert_emiss.exe[0x24e5770]
./convert_emiss.exe[0x20e6423]
./convert_emiss.exe[0x4077aa]
./convert_emiss.exe[0x406f71]
/lib64/libc.so.6(__libc_start_main+0xf5)[0x2b3ff9ff23d5]
./convert_emiss.exe[0x406fa1]
======= Memory map: ========
……
2b3ffc7bc000-2b3ffc806000 r-xp 00000000 fd:00 134257705 /usr/lib64/libgssapi_krb5.so.2.2
2b3ffc806000-2b3ffca06000 ---p 0004a000 fd:00 134257705 /usr/lib64/libgssapi_krb5.so.2.2
2b3ffca06000-2b3ffca07000 r--p 0004a000 fd:00 134257705 /usr/lib64/libgssapi_krb5.so.2.2
2b3ffca07000-2b3ffca09000 rw-p 0004b000 fd:00 134257705 /usr/lib64/libgssapi_krb5.so.2.2
2b3ffca09000-2b3ffcae2000 r-xp 00000000 fd:00 134257715 /usr/lib64/libkrb5.so.3.3
Program received signal SIGABRT: Process abort signal.
Backtrace for this error:
#0 0x2B3FF9573697
#1 0x2B3FF9573CDE
#2 0x2B3FFA00627F
#3 0x2B3FFA006207
#4 0x2B3FFA0078F7
#5 0x2B3FFA048D26
#6 0x2B3FFA04F5D3
#7 0x2B3FFA052D0F
#8 0x2B3FFA0551AB
#9 0x36AAC27 in __module_ra_cam_support_MOD_oznini
#10 0x2896072 in __module_physics_init_MOD_ra_init
#11 0x28BAAE7 in __module_physics_init_MOD_phy_init
#12 0x253F808 in start_domain_em_
#13 0x24E576F in start_domain_
#14 0x20E6422 in med_initialdata_input_
#15 0x4077A9 in MAIN__ at convert_emiss.f90:?
Could anyone please share the methods of solving this problem? That will be really helpful for me. Thanks.