如何解决通过 Torque 将 Julia 文件提交到集群时导入 PyPlot 的问题
我正在学校的集群上运行一些 Julia 代码。调用 test_julia.jl 的 .bps 文件如下：
1 #!/bin/tcsh
2
3 #PBS -l walltime=24:00:00
4 #PBS -l nodes=1:ppn=1
5 #PBS -N test_julia
6 ##PBS -q normal
7 #PBS -q normal
8 #PBS -j oe
9 #PBS -o ./log/julia.${PBS_JOBID}.out
10 ##PBS -e ./log/julia.${PBS_JOBID}.err
11 ##PBS -m abe -M myemail@school.edu
12 #PBS -t 1-38
13
14
15 cd $PBS_O_WORKDIR
16
17 module load julia
18
19 julia /path/to/file/test_julia.jl 10 20 30
20
其中 10、20 和 30 是 test_julia.jl 的输入。当我自己运行这个 .jl 文件时，它本身运行良好。在 .jl 文件中，我调用了
1 import Distributed
2 using Distributed
3
4 @everywhere pushfirst!(Base.DEPOT_PATH,"/tmp/test.cache")
开头。
当我使用 Torque 提交上述文件时，.out 文件中出现以下错误：
1 Warning: no access to tty (Bad file descriptor).
2 Thus no job control in this shell.
3
4 signal (6): Aborted
5 in expression starting at /data/heathjo/.julia/packages/PyPlot/XHEG0/src/colormaps.jl:4
6 gsignal at /lib64/libc.so.6 (unknown line)
7 abort at /lib64/libc.so.6 (unknown line)
8 jl_intref at /buildworker/worker/package_linux64/build/src/typemap.c:190 [inlined]
9 mtcache_hash_lookup at /buildworker/worker/package_linux64/build/src/typemap.c:251 [inlined]
10 jl_typemap_intersection_visitor at /buildworker/worker/package_linux64/build/src/typemap.c:526
11 jl_method_table_insert at /buildworker/worker/package_linux64/build/src/gf.c:1622
12 _jl_restore_incremental at /buildworker/worker/package_linux64/build/src/dump.c:2174
13 jl_restore_incremental at /buildworker/worker/package_linux64/build/src/dump.c:3159
14 _include_from_serialized at ./loading.jl:617
15 _require_search_from_serialized at ./loading.jl:713
16 _require at ./loading.jl:937
17 require at ./loading.jl:858
18 require at ./loading.jl:853
19 jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2219
20 jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1571 [inlined]
21 call_require at /buildworker/worker/package_linux64/build/src/toplevel.c:397 [inlined]
22 eval_import_path at /buildworker/worker/package_linux64/build/src/toplevel.c:432
23 jl_toplevel_eval_flex at /buildworker/worker/package_linux64/build/src/toplevel.c:630
24 jl_parse_eval_all at /buildworker/worker/package_linux64/build/src/ast.c:883
25 jl_load at /buildworker/worker/package_linux64/build/src/toplevel.c:826
26 include at ./boot.jl:326 [inlined]
27 include_relative at ./loading.jl:1038
28 include at ./sysimg.jl:29 [inlined]
29 include at /data/heathjo/.julia/packages/PyPlot/XHEG0/src/PyPlot.jl:2
30 jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2219
31 do_call at /buildworker/worker/package_linux64/build/src/interpreter.c:323
32 eval_value at /buildworker/worker/package_linux64/build/src/interpreter.c:411
33 eval_stmt_value at /buildworker/worker/package_linux64/build/src/interpreter.c:362 [inlined]
34 eval_body at /buildworker/worker/package_linux64/build/src/interpreter.c:773
35 jl_interpret_toplevel_thunk_callback at /buildworker/worker/package_linux64/build/src/interpreter.c:885
36 unknown function (ip: 0xfffffffffffffffe)
37 unknown function (ip: 0x2aaab83d046f)
38 unknown function (ip: 0xffffffffffffffff)
39 jl_interpret_toplevel_thunk at /buildworker/worker/package_linux64/build/src/interpreter.c:894
40 jl_toplevel_eval_flex at /buildworker/worker/package_linux64/build/src/toplevel.c:764
41 jl_eval_module_expr at /buildworker/worker/package_linux64/build/src/toplevel.c:179
42 jl_toplevel_eval_flex at /buildworker/worker/package_linux64/build/src/toplevel.c:614
43 jl_parse_eval_all at /buildworker/worker/package_linux64/build/src/ast.c:883
44 jl_load at /buildworker/worker/package_linux64/build/src/toplevel.c:826
45 include at ./boot.jl:326 [inlined]
46 include_relative at ./loading.jl:1038
47 include at ./sysimg.jl:29
48 jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2219
49 do_call at /buildworker/worker/package_linux64/build/src/interpreter.c:323
50 eval_value at /buildworker/worker/package_linux64/build/src/interpreter.c:411
51 eval_stmt_value at /buildworker/worker/package_linux64/build/src/interpreter.c:362 [inlined]
52 eval_body at /buildworker/worker/package_linux64/build/src/interpreter.c:773
53 eval_body at /buildworker/worker/package_linux64/build/src/interpreter.c:689
54 jl_interpret_toplevel_thunk_callback at /buildworker/worker/package_linux64/build/src/interpreter.c:885
55 unknown function (ip: 0xfffffffffffffffe)
56 unknown function (ip: 0x2aaab8b7761f)
57 unknown function (ip: 0x2)
58 jl_interpret_toplevel_thunk at /buildworker/worker/package_linux64/build/src/interpreter.c:894
59 jl_toplevel_eval_flex at /buildworker/worker/package_linux64/build/src/toplevel.c:764
60 jl_toplevel_eval_in at /buildworker/worker/package_linux64/build/src/toplevel.c:793
61 eval at ./boot.jl:328 [inlined]
62 eval at ./client.jl:404
63 jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2219
64 top-level scope at ./none:3
65 jl_fptr_trampoline at /buildworker/worker/package_linux64/build/src/gf.c:1864
66 jl_toplevel_eval_flex at /buildworker/worker/package_linux64/build/src/toplevel.c:758
67 jl_toplevel_eval_in at /buildworker/worker/package_linux64/build/src/toplevel.c:793
68 eval at ./boot.jl:328
69 jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2219
70 exec_options at ./client.jl:243
71 _start at ./client.jl:436
72 jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2219
73 jl_apply at /buildworker/worker/package_linux64/build/ui/../src/julia.h:1571 [inlined]
74 true_main at /buildworker/worker/package_linux64/build/ui/repl.c:96
75 main at /buildworker/worker/package_linux64/build/ui/repl.c:217
76 __libc_start_main at /lib64/libc.so.6 (unknown line)
77 _start at /gsfs0/public/julia/julia-1.1.0/bin/julia (unknown line)
78 Allocations: 7287740 (Pool: 7286538; Big: 1202); GC: 15
79 ERROR: LoadError: Failed to precompile PyPlot [d330b81b-6aea-500a-939a-2ce795aea3ee] to /tmp/test.cache/compiled/v1.1/PyPlot/oatAj.ji.
80 error(::String) at ./error.jl:33
81 compilecache(::Base.PkgId,::String) at ./loading.jl:1197
82 _require(::Base.PkgId) at ./loading.jl:960
83 require(::Base.PkgId) at ./loading.jl:858
84 require(::Module,::Symbol) at ./loading.jl:853
85 eval at ./boot.jl:328 [inlined]
86 (::getfield(Distributed,Symbol("##164#166")){Module,Expr})() at ./task.jl:259
87 Stacktrace:
88 [1] sync_end(::Array{Any,1}) at ./task.jl:226
89 [2] macro expansion at ./task.jl:245 [inlined]
90 [3] remotecall_eval(::Module,::Array{Int64,1},::Expr) at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.1/Distributed/src/macros.jl:206
91 [4] top-level scope at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.1/Distributed/src/macros.jl:190
92 [5] include at ./boot.jl:326 [inlined]
93 [6] include_relative(::Module,::String) at ./loading.jl:1038
94 [7] include(::Module,::String) at ./sysimg.jl:29
95 [8] exec_options(::Base.JLOptions) at ./client.jl:267
96 [9] _start() at ./client.jl:436
97 in expression starting at /gsfs0/data/heathjo/Projects/Phase_Transitions_in_QA_Circuits/Acceptable_Code/Entropy_vs_L_and_p/Entropy_Plotter.jl:7
98 ssh: No match.
我真的不知道如何处理这个错误，或者我应该如何修改我的 .bps 文件来修复它。它似乎与 PyPlot（我在 test_julia.jl 中调用它）有关，但是如果我直接运行 test_julia.jl，我永远不会收到此错误。我也只在调用 PyPlot 时收到此错误，而我事先在 .jl 文件中调用的其他包都没有。有什么简单的方法可以解决这个问题吗？
版权声明：本文内容由互联网用户自发贡献，该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容， 请发送邮件至 dio@foxmail.com 举报，一经查实，本站将立刻删除。