January 2021 - Compass-ci - mailweb.openeuler.org

[PATCH v3 compass-ci] create-image: disable selinux
by Wang Chenglong 21 Jan '21

21 Jan '21

[Error] [ 5.786497] Run /sbin/init as init process SELinux: Could not open policy file <= /etc/selinux/targeted/policy/policy.32: No such file or directory Signed-off-by: Wang Chenglong <18509160991(a)163.com> --- container/osimage/openeuler/create-image | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/container/osimage/openeuler/create-image b/container/osimage/openeuler/create-image index 631bb14..e161b17 100755 --- a/container/osimage/openeuler/create-image +++ b/container/osimage/openeuler/create-image @@ -13,6 +13,17 @@ setup_rootfs() ln -fs /usr/share/zoneinfo/Asia/Shanghai /etc/localtime rm -f /etc/machine-id /var/lib/dbus/machine-id systemd-machine-id-setup + + # disable selinux + echo "Disable selinux" + if [ -f /etc/selinux/config ]; then + sed -i -r 's/^.*SELINUX=enforcing.*$/SELINUX=disabled/g' "/etc/selinux/config" + else + cat > "/etc/selinux/config" <<-EOF + SELINUX=disabled + SELINUXTYPE=targeted + EOF + fi } setup_login() { -- 2.23.0

1 1

[PATCH v2 lkp-tests] pkg: add PKGBUILD for openeuler docker test
by Bai Jing 20 Jan '21

20 Jan '21

[why] The source file is missing in the openEuler image, So we need to add source files for it. Signed-off-by: Bai Jing <799286817(a)qq.com> --- pkg/openeuler_docker/PKGBUILD | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) create mode 100644 pkg/openeuler_docker/PKGBUILD diff --git a/pkg/openeuler_docker/PKGBUILD b/pkg/openeuler_docker/PKGBUILD new file mode 100644 index 000000000..6a1190718 --- /dev/null +++ b/pkg/openeuler_docker/PKGBUILD @@ -0,0 +1,27 @@ +pkgname=openeuler_docker +pkgver=1.0 +pkgrel=1 +pkgdesc="openeuler docker image" +arch=('i386' 'x86_64') +url="https://repo.openeuler.org/openEuler-20.03-LTS/docker_img" +license=('GPL') +source=("https://repo.openeuler.org/openEuler-20.03-LTS/docker_img/aarch64/openEuler…") +md5sums=('6aa300910c9f9040b725939fb80e8121') + +handle_docker_image() +{ + docker load < openEuler-docker.aarch64.tar.xz + docker run -d --name my-openeuler openeuler-20.03-lts:latest + container_id=$(docker ps -aqf 'name=my-openeuler') + docker cp /etc/yum.repos.d/openEuler.repo $container_id:/etc/yum.repos.d/ + docker commit $container_id openeuler-20.03-lts:v1 + docker save -o openeuler-20.03-lts-v1.tar openeuler-20.03-lts:v1 + docker rm -f $container_id +} + +package() +{ + handle_docker_image + mkdir -p "$pkgdir/lkp/benchmarks/${pkgname}" + cp "$srcdir/openeuler-20.03-lts-v1.tar" "$pkgdir/lkp/benchmarks/${pkgname}" +} -- 2.23.0

1 0

[PATCH v4 compass-ci] git_mirror.rb: fix that es-states clean when git-mirror restart
by Li Yuanchao 20 Jan '21

20 Jan '21

Everytime git-mirror get new_refs or just git fetch, these actions and the time will be recorded in hash fork_stat{}. But when container git-mirror restart, fork_stat{} will be initialized. So historical data can't be kept. Now, git-mirror will get historical data from es when initialize hash fork_stat{}. So historical data will be kept. Signed-off-by: Li Yuanchao <lyc163mail(a)163.com> --- lib/git_mirror.rb | 30 +++++++++++++++++++++--------- 1 file changed, 21 insertions(+), 9 deletions(-) diff --git a/lib/git_mirror.rb b/lib/git_mirror.rb index 86f02e5..4f74751 100644 --- a/lib/git_mirror.rb +++ b/lib/git_mirror.rb @@ -109,15 +109,8 @@ class MirrorMain @webhook_queue = channel.queue('web_hook') end - def fork_stat_init(stat_key) - @fork_stat[stat_key] = { - queued: false, - priority: 0, - fetch_time: [], - offset_fetch: 0, - new_refs_time: [], - offset_new_refs: 0 - } + def fork_stat_init(git_repo) + @fork_stat[git_repo] = get_fork_stat(git_repo) end def load_defaults(repodir) @@ -372,4 +365,23 @@ class MirrorMain handle_submodule(submodule) end + + def get_fork_stat(git_repo) + fork_stat = { + queued: false, + priority: 0, + fetch_time: [], + offset_fetch: 0, + new_refs_time: [], + offset_new_refs: 0 + } + query = { query: { match: { git_repo: git_repo } } } + result = @es_client.search(index: 'repo', body: query)['hits'] + return fork_stat unless result['total'].positive? + + fork_stat.each_key do |key| + fork_stat[key] = result['hits'][0]['_source'][key.to_s] + end + return fork_stat + end end -- 2.23.0

1 0

[PATCH v3 compass-ci] git_mirror.rb: fix that es-states clean when git-mirror restart
by Li Yuanchao 20 Jan '21

20 Jan '21

Everytime git-mirror get new_refs or just git fetch, these actions and the time will be recorded in hash fork_stat{}. But when container git-mirror restart, fork_stat{} will be initialized. So historical data can't be kept. Now, git-mirror will get historical data from es when initialize hash fork_stat{}. So historical data will be kept. Signed-off-by: Li Yuanchao <lyc163mail(a)163.com> --- lib/git_mirror.rb | 35 ++++++++++++++++++++++++++--------- 1 file changed, 26 insertions(+), 9 deletions(-) diff --git a/lib/git_mirror.rb b/lib/git_mirror.rb index 86f02e5..6bb474a 100644 --- a/lib/git_mirror.rb +++ b/lib/git_mirror.rb @@ -109,15 +109,8 @@ class MirrorMain @webhook_queue = channel.queue('web_hook') end - def fork_stat_init(stat_key) - @fork_stat[stat_key] = { - queued: false, - priority: 0, - fetch_time: [], - offset_fetch: 0, - new_refs_time: [], - offset_new_refs: 0 - } + def fork_stat_init(git_repo) + @fork_stat[git_repo] = get_fork_stat(git_repo) end def load_defaults(repodir) @@ -372,4 +365,28 @@ class MirrorMain handle_submodule(submodule) end + + def copy_from_es(es_data) + fork_stat = {} + es_data.each do |key, value| + fork_stat[key.to_sym] = value + end + return fork_stat + end + + def get_fork_stat(git_repo) + query = { query: { match: { git_repo: git_repo } } } + result = @es_client.search(index: 'repo', body: query)['hits'] + return copy_from_es(result['hits'][0]['_source']) if result['total'].positive? + + fork_stat = { + queued: false, + priority: 0, + fetch_time: [], + offset_fetch: 0, + new_refs_time: [], + offset_new_refs: 0 + } + return fork_stat + end end -- 2.23.0

1 0

[PATCH v3 compass-ci 1/2] qemu/kvm.sh: optimize the prompt when the error occurs
by Yu Chuan 20 Jan '21

20 Jan '21

Signed-off-by: Yu Chuan <13186087857(a)163.com> --- providers/qemu/kvm.sh | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git a/providers/qemu/kvm.sh b/providers/qemu/kvm.sh index 35678fbd44ef..93977ec15438 100755 --- a/providers/qemu/kvm.sh +++ b/providers/qemu/kvm.sh @@ -6,6 +6,9 @@ : ${nr_cpu:=1} : ${memory:=1G} +: ${LKP_SRC:="/c/lkp-tests"} + +source ${LKP_SRC}/lib/log.sh check_logfile() { @@ -52,15 +55,23 @@ parse_ipxe_script() check_option_value() { + [ -n "$kernel" ] || { + log_info "Can not find job for current hostname: $hostname." + exit 0 + } + [ -s "$kernel" ] || { - echo "The kernel does not exist: $kernel" - exit + log_error "Can not find kernel file or kernel file is empty: $kernel." + exit 1 } # debian has both qemu-system-x86_64 and qemu-system-riscv64 command [[ $kernel =~ 'riscv64' ]] && qemu=qemu-system-riscv64 - [ -n "$initrds" ] || exit + [ -n "$initrds" ] || { + log_error "Can not find any initrd for current vm." + exit 1 + } } set_initrd() -- 2.23.0

3 3

[PATCH v2 compass-ci 2/2] service/scheduler: save testbox info in es
by Wu Zhende 20 Jan '21

20 Jan '21

When a testbox get a job, record its information in es. The deadline field indicates the time when the job executed by the testbox times out. It's a sign that other measures are needed to intervene. Can use curl -XGET http://localhost:9200/testbox/_doc/_search -d '{"query":{"range":{"deadline":{"lt":"now"}}}}' to find all timeout testboxs and jobs Signed-off-by: Wu Zhende <wuzhende666(a)163.com> --- src/lib/sched.cr | 19 +++++++++++++++++++ src/scheduler/find_job_boot.cr | 1 + 2 files changed, 20 insertions(+) diff --git a/src/lib/sched.cr b/src/lib/sched.cr index 51e0d74..8d7c218 100644 --- a/src/lib/sched.cr +++ b/src/lib/sched.cr @@ -134,6 +134,25 @@ class Sched @log.warn(e) end + def set_tbox_boot_wtmp(job : Job) + time = Time.local + booting_time = time.to_s("%Y-%m-%dT%H:%M:%S") + + runtime = (job["timeout"]? || job["runtime"]?).to_s + runtime = 1800 if runtime.empty? + + # reserve 300 seconds for system startup, hw machine will need such long time + deadline = (time + (runtime.to_i32 * 2 + 300).second).to_s("%Y-%m-%dT%H:%M:%S") + hash = { + "job_id" => job["id"], + "state" => "booting", + "booting_time" => booting_time, + "deadline" => deadline + } + + @es.update_tbox(job["testbox"], hash) + end + def report_ssh_port testbox = @env.params.query["tbox_name"] ssh_port = @env.params.query["ssh_port"].to_s diff --git a/src/scheduler/find_job_boot.cr b/src/scheduler/find_job_boot.cr index b7f3d75..f8d7caa 100644 --- a/src/scheduler/find_job_boot.cr +++ b/src/scheduler/find_job_boot.cr @@ -101,6 +101,7 @@ class Sched if job create_job_cpio(job.dump_to_json_any, Kemal.config.public_folder) + set_tbox_boot_wtmp(job) else # for physical machines spawn { auto_submit_idle_job(host) } -- 2.23.0

1 0

[PATCH v2 compass-ci 1/2] scheduler/elasticsearch_client.cr: support update all hash
by Wu Zhende 20 Jan '21

20 Jan '21

[Before] Only support update job_id/job_state/history [After] Support update any key-value. Because need save other new key-value in testbox. Signed-off-by: Wu Zhende <wuzhende666(a)163.com> --- src/scheduler/elasticsearch_client.cr | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/src/scheduler/elasticsearch_client.cr b/src/scheduler/elasticsearch_client.cr index b17d7f8..c1663c9 100644 --- a/src/scheduler/elasticsearch_client.cr +++ b/src/scheduler/elasticsearch_client.cr @@ -3,6 +3,7 @@ require "yaml" require "json" +require "any_merge" require "elasticsearch-crystal/elasticsearch/api" require "./constants" require "../lib/job" @@ -89,21 +90,30 @@ class Elasticsearch::Client def update_tbox(testbox : String, wtmp_hash : Hash) query = {:index => "testbox", :type => "_doc", :id => testbox} - result = @client.get_source(query) if @client.exists(query) - history = result["history"].as_a? if result.is_a?(JSON::Any) + if @client.exists(query) + result = @client.get_source(query) + raise result unless result.is_a?(JSON::Any) + + history = result["history"].as_a? + result = result.as_h + else + result = wtmp_hash + end + history ||= [] of JSON::Any history << JSON.parse(wtmp_hash.to_json) + history = JSON.parse(history.to_json) + + body = { "history" => history} + body.any_merge!(result) + body.any_merge!(wtmp_hash) @client.create( { :index => "testbox", :type => "_doc", :id => testbox, - :body => { - :state => wtmp_hash["state"], - :job_id => wtmp_hash["job_id"], - :history => history - } + :body => body } ) end -- 2.23.0

1 0

[PATCH compass-ci] providers/qemu: fix script hangs when no job now
by Liu Yinsi 20 Jan '21

20 Jan '21

[error] when no job now, $initrds is null, execute my-qemu.sh will hangs. lys@crystal ~/compass-ci/providers% ./my-qemu.sh hostname: vm-2p8g.lys-3917841 mac: 0a-4b-33-a0-96-69 DoneDone % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 79 100 79 0 0 19750 0 --:--:-- --:--:-- --:--:-- 19750 lys@crystal ~/compass-ci/providers/qemu% cat --help cat Copy standard input to standard output. when input is null, script will hang to waiting for input. [how] function set_options(): cat $initrds > $initrd before: execut cat command, then check after: check kernel, check initrds, if not null, execute cat command. in addition, check_option_value -> check_kernel, check_initrd, check_qemu convenient to call as needed. Signed-off-by: Liu Yinsi <liuyinsi(a)163.com> --- providers/qemu/kvm.sh | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/providers/qemu/kvm.sh b/providers/qemu/kvm.sh index 4d05964..b4e87fc 100755 --- a/providers/qemu/kvm.sh +++ b/providers/qemu/kvm.sh @@ -53,7 +53,7 @@ parse_ipxe_script() done < $ipxe_script } -check_option_value() +check_kernel() { [ -n "$kernel" ] || { log_info "Can not find job for current hostname: $hostname." @@ -64,10 +64,16 @@ check_option_value() log_error "Can not find kernel file or kernel file is empty: $kernel." exit 1 } - +} + +check_qemu() +{ # debian has both qemu-system-x86_64 and qemu-system-riscv64 command [[ $kernel =~ 'riscv64' ]] && qemu=qemu-system-riscv64 +} +check_initrds() +{ [ -n "$initrds" ] || { log_error "The current initrds is null." exit 1 @@ -118,6 +124,8 @@ set_qemu() do command -v "$qemu" > /dev/null && break done + + check_qemu } print_message() @@ -209,8 +217,10 @@ write_logfile parse_ipxe_script +check_kernel +check_initrds + set_options -check_option_value print_message -- 2.23.0

1 1

[PATCH v11 compass-ci] lib/es_jobs.rb: optimize es-jobs output
by Lu Kaiyi 20 Jan '21

20 Jan '21

[why] Reduce the nesting of results and output yaml [example] es-jobs os=archlinux [output] kvcount.arch=aarch64: 43 kvcount.arch=x86_64: 2 kvcount.category=benchmark: 12 kvcount.category=functional: 33 kvcount.job_state=failed: 20 kvcount.job_state=finished: 25 kvcount.os=archlinux: 45 kvcount.suite=cci-depends: 28 kvcount.suite=cci-makepkg: 5 kvcount.suite=iperf: 3 kvcount.suite=iperf-walk-os-test: 9 kvcount.summary.any_error=1: 24 kvcount.summary.any_stderr=1: 44 kvcount.summary.any_warning=1: 23 kvcount.summary.success=1: 1 kvcount.tbox_group=vm-2p16g--wangyong: 1 kvcount.tbox_group=vm-2p16g--wcl: 1 kvcount.tbox_group=vm-2p16g.wangyong: 34 kvcount.tbox_group=vm-2p8g: 9 ... raw.id.[job_state=failed]: ["crystal.122405","crystal.122406","crystal.122331","crystal.122365","crystal.122404","crystal.122438","crystal.122607","crystal.122370","crystal.122375","crystal.122410","crystal.122900","crystal.122906","crystal.122271","crystal.122451","crystal.122835","crystal.122834","crystal.122232","crystal.122403","crystal.122453","crystal.122886"] ... raw.id.[summary.any_error=1]: ["crystal.122436","crystal.122476","crystal.122803","crystal.122883","crystal.122910","crystal.122438","crystal.122452","crystal.122895","crystal.122446","crystal.122685","crystal.122687","crystal.122833","crystal.122451","crystal.122690","crystal.122719","crystal.122882","crystal.122890","crystal.122453","crystal.122443","crystal.122435","crystal.122432","crystal.122688","crystal.122689","crystal.122886"] ... sum.stats.last_state.test.iperf.exit_code.127: 12 sum.stats.stderr./lkp/lkp/src/lib/env.sh:#:set_perf_path: 12 sum.stats.last_state.is_incomplete_run: 12 sum.stats.kmsg.timestamp:last: 45 sum.stats.last_state.exit_fail: 20 sum.stats.stderr.has_stderr: 44 ... raw.stats.sched_debug.cfs_rq:/.load.stddev: [524288.0,null,null,null,null,null,null,null,null,524288.0,524288.0,null,null,null,null,null,516608.0,null,1048576.0,null,null,null,null,null,null,null,524288.0,null,null,null,null,2104832.0,1572864.0,null,null,2097152.0,null,null,null,null,null,null,null,null,null] ... avg.stats.sched_debug.cfs_rq:/.load.stddev: 1048576.0 avg.stats.softirqs.CPU1.NET_RX: 2.5833333333333335 avg.stats.slabinfo.kmalloc-512.active_objs: 1372.75 ... max.stats.sched_debug.cfs_rq:/.load.stddev: 2104832.0 max.stats.softirqs.CPU1.NET_RX: 6 max.stats.slabinfo.kmalloc-512.active_objs: 1616 ... Signed-off-by: Lu Kaiyi <2392863668(a)qq.com> --- lib/es_jobs.rb | 202 +++++++++++++++++++++++++++---------------------- 1 file changed, 112 insertions(+), 90 deletions(-) diff --git a/lib/es_jobs.rb b/lib/es_jobs.rb index a90b1ba..94c3193 100644 --- a/lib/es_jobs.rb +++ b/lib/es_jobs.rb @@ -3,137 +3,159 @@ LKP_SRC = ENV['LKP_SRC'] || '/c/lkp-tests' +KEYWORD = %w[suite os arch category job_state tbox_group upstream_repo summary.success + summary.any_fail summary.any_error summary.any_stderr summary.any_warning].freeze + +require 'json' require "#{LKP_SRC}/lib/stats" require_relative './es_query' # deal jobs search from es class ESJobs def initialize(es_query, my_refine = [], fields = [], stats_filter = []) - @es_query = es_query - @es = ESQuery.new(ES_HOST, ES_PORT) + @jobs = query_jobs_from_es(es_query) @refine = my_refine @fields = fields @stats_filter = stats_filter - @stats_filter_result = {} @refine_jobs = [] - @jobs = {} - @stats_level = { - 0 => 'stats.success', - 1 => 'stats.unknown', - 2 => 'stats.warning', - 3 => 'stats.has_error' - } - set_defaults - deal_jobs + set_jobs_summary end - def set_defaults - query_result = @es.multi_field_query(@es_query) - query_result['hits']['hits'].each do |job| - @jobs[job['_id']] = job['_source'] - end - - @stats = { - 'stats.count' => Hash.new(0), - 'stats.sum' => Hash.new(0), - 'stats.avg' => Hash.new(0) - } - @result = {} - @fields.each do |field| - @result[field] = [] - end + def query_jobs_from_es(items) + es = ESQuery.new(ES_HOST, ES_PORT) + result = es.multi_field_query items + jobs = result['hits']['hits'] + jobs.map! { |job| job['_source'] } + return jobs end - def add_result_fields(job, level) - return unless @refine.include?(level) || @refine.include?(-1) - - @refine_jobs << job['id'] - @fields.each do |field| - value = job[field] - if value - value = job['id'] + '.' + value if field == 'job_state' - @result[field] << value + def set_job_summary(stats, job) + summary_result = '' + stats.keys.each do |stat| + if stat.match(/\.warning/i) + job['summary.any_warning'] = 1 + summary_result = 'warning' + end + if stat.match(/stderr\./i) + job['summary.any_stderr'] = 1 + summary_result = 'stderr' + end + if stat.match(/error/i) + job['summary.any_error'] = 1 + summary_result = 'error' + end + if stat.match(/\.fail$/i) + job['summary.any_fail'] = 1 + summary_result = 'fail' end - - next unless job['stats'] - - @result[field] << job['stats'][field] if job['stats'][field] end - end - - def deal_jobs - stats_count = Hash.new(0) - stats_jobs = {} + return unless summary_result.empty? - @jobs.each do |job_id, job| - level = deal_stats(job) - add_result_fields(job, level) + job['summary.success'] = 1 + end - stat_key = @stats_level[level] - stat_jobs_key = stat_key + '_jobs' + # set jobs summary fields information in place + def set_jobs_summary + @jobs.each do |job| + stats = job['stats'] + next unless stats - stats_count[stat_key] += 1 - stats_jobs[stat_jobs_key] ||= [] - stats_jobs[stat_jobs_key] << job_id + set_job_summary(stats, job) end - - @stats['stats.count'].merge!(stats_count) - @stats['stats.count'].merge!(stats_jobs) end - def deal_stats(job, level = 0) - return 1 unless job['stats'] + def get_all_metrics(jobs) + metrics = [] + jobs.each do |job| + stats = job['stats'] + next unless stats - job['stats'].each do |key, value| - match_stats_filter(key, value, job['id']) - calculate_stat(key, value) - level = get_stat_level(key, level) + metrics.concat(stats.keys) end - return level + metrics.uniq! end - def match_stats_filter(key, value, job_id) - @stats_filter.each do |filter| - next unless key.include?(filter) + def initialize_result_hash(metrics) + result = { + 'kvcount' => {}, + 'raw.id' => {}, + 'sum.stats' => {}, + 'raw.stats' => {}, + 'avg.stats' => {}, + 'max.stats' => {}, + 'min.stats' => {} + } + metrics.each { |metric| result['raw.stats'][metric] = [] } + result + end - key = job_id + '.' + key - @stats_filter_result[key] = value + def set_default_value(result, stats, metrics) + left_metrics = metrics - stats.keys + left_metrics.each { |metric| result['raw.stats'][metric] << nil } - break + stats.each do |key, value| + result['raw.stats'][key] << value end end - def calculate_stat(key, value) - if function_stat?(key) - return unless @fields.include?('stats.sum') + def kvcount(result, job) + KEYWORD.each do |keyword| + next unless job[keyword] - @stats['stats.sum'][key] += value - else - return unless @fields.include?('stats.avg') + result['kvcount']["#{keyword}=#{job[keyword]}"] ||= 0 + result['kvcount']["#{keyword}=#{job[keyword]}"] += 1 + result['raw.id']["[#{keyword}=#{job[keyword]}]"] ||= [] + result['raw.id']["[#{keyword}=#{job[keyword]}]"] << job['id'] + end + end - @stats['stats.avg'][key] = (@stats['stats.avg'][key] + value) / 2 + def stats_count(result) + result['raw.stats'].each do |key, value| + if function_stat?(key) + result['sum.stats'][key] = value.compact.size + else + result['avg.stats'][key] = value.compact.sum / value.compact.size.to_f + result['max.stats'][key] = value.compact.max + result['min.stats'][key] = value.compact.min + end end end - def get_stat_level(stat, level) - return level if level >= 3 - return 3 if stat.match(/error|fail/i) - return 2 if stat.match(/warn/i) + def query_jobs_state(jobs) + metrics = get_all_metrics(jobs) + result = initialize_result_hash(metrics) + jobs.each do |job| + stats = job['stats'] + next unless stats + + set_default_value(result, stats, metrics) + kvcount(result, job) + end - return 0 + stats_count(result) + result end - def output - result = { - 'stats.count' => @stats['stats.count'] - } + def output_yaml(prefix, result) + result.each do |key, value| + if prefix.empty? + prefix_key = "#{key}" + else + prefix_key = "#{prefix}.#{key}" + end - @stats.each do |key, value| - result[key] = value if @fields.include?(key) + if value.is_a? Hash + output_yaml(prefix_key, value) + else + puts "#{prefix_key}: #{value.to_json}" + end end + end - @result['stats_filter_result'] = @stats_filter_result unless @stats_filter.empty? - @result.merge!(result) - puts JSON.pretty_generate(@result) + def output + @result = query_jobs_state(@jobs) + @result['kvcount'] = @result['kvcount'].sort.to_h + @result['raw.id'] = @result['raw.id'].sort.to_h + output_yaml('', @result) end end -- 2.23.0

1 0

[PATCH v5 compass-ci 1/2] providers/qemu: fix script hangs when no job now
by Liu Yinsi 20 Jan '21

20 Jan '21

[error] when no job now, $initrds is null, execute my-qemu.sh will hangs. lys@crystal ~/compass-ci/providers% ./my-qemu.sh hostname: vm-2p8g.lys-3917841 mac: 0a-4b-33-a0-96-69 DoneDone % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 79 100 79 0 0 19750 0 --:--:-- --:--:-- --:--:-- 19750 lys@crystal ~/compass-ci/providers/qemu% cat --help cat Copy standard input to standard output. when input is null, script will hang to waiting for input. [how] before: execute cat command, then judge $initrds is null or not after: judge $initrds is null or not, if not null, execute cat command Signed-off-by: Liu Yinsi <liuyinsi(a)163.com> --- providers/qemu/kvm.sh | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/providers/qemu/kvm.sh b/providers/qemu/kvm.sh index 4d05964..0c58d94 100755 --- a/providers/qemu/kvm.sh +++ b/providers/qemu/kvm.sh @@ -68,16 +68,17 @@ check_option_value() # debian has both qemu-system-x86_64 and qemu-system-riscv64 command [[ $kernel =~ 'riscv64' ]] && qemu=qemu-system-riscv64 - [ -n "$initrds" ] || { + if [ -n "$initrds" ]; then + cat $initrds > $initrd + else log_error "The current initrds is null." exit 1 - } + fi } set_initrd() { initrd=initrd - cat $initrds > $initrd } set_bios() -- 2.23.0

2 2