我有两个 sh 脚本，分别称为 Z-backup 和 lb。Z-backup 是我放在 cron.daily 中的脚本，因此它作为 anacron 作业运行并调用 lb 脚本。Z-backup 脚本的内容为：
#!/bin/sh
set -e
# Anacron script to run a daily backup at boot time.
exec /home/chh1/bin/lb >/dev/null 2>&1
exit
lb脚本的内容为：
#!/bin/sh
# Script to do a backup of PostgreSQL database crewdb and /home folder.
# ~/.pgpass file has been created to support pg_dump -w flag.
# The following if statement removes crewdb.backup.sql if it exists.
if [ -e /home/chh1/Dropbox/postgres_backup/crewdb.backup.sql ]
then
rm /home/chh1/Dropbox/postgres_backup/crewdb.backup.sql
fi >/dev/null 2>&1
echo "PostgreSQL database crewdb backed up on $(date)" >> /home/chh1
/logs/postgres_bak_log.txt
pg_dump -h localhost -p 5432 -U postgres -d crewdb -w -C -F p -b -v -f
~/Dropbox/postgres_backup/crewdb.backup.sql >/dev/null 2>&1
echo "Home folder backed up on $(date)" >> /home/chh1
/logs/home_bak_log.txt
rsync -acu --delete --exclude-from='/home/chh1/rsync_exclude.txt' /home/
/media/chh1/f86f92f7-bcb4-4b8e-8a64-8a210606d422/home_backup/ >/dev/null
2>&1
if [ -e /media/chh1/f86f92f7-bcb4-4b8e-8a64-8a210606d422/home_backup
/chh1/Dropbox/postgres_backup/crewdb.backup.sql ]
then
rm /media/chh1/f86f92f7-bcb4-4b8e-8a64-8a210606d422/home_backup
/chh1/Dropbox/postgres_backup/crewdb.backup.sql
fi >/dev/null 2>&1
pg_dump -h localhost -p 5432 -U postgres -d crewdb -w -C -F p -b -v -f
/media/chh1/f86f92f7-bcb4-4b8e-8a64-8a210606d422/home_backup/chh1/Dropbox
/postgres_backup/crewdb.backup.sql >/dev/null 2>&1
exit
如果我只是在终端中运行 lb，脚本就会按照我想要的方式正常工作。（运行 lb 后，我运行 echo $? 并获得退出状态 0）。但是，计划的 anacron 作业仅部分执行，我无法弄清楚为什么会这样。日志条目已创建，rsync 命令有效，因此主文件夹是通过 anacron 而不是 pg_dump 备份的。当我在 lb 中运行 pg_dump 命令并将 stderr 保存到文件时，例如：
pg_dump -h localhost -p 5432 -U postgres -d crewdb -w -C -F p -b -v -f
~/Dropbox/postgres_backup/crewdb.backup.sql 2>/pg_dump_errors.txt
和
pg_dump -h localhost -p 5432 -U postgres -d crewdb -w -C -F p -b -v -f
/media/chh1/f86f92f7-bcb4-4b8e-8a64-8a210606d422/home_backup/chh1/Dropbox
/postgres_backup/crewdb.backup.sql 2>/pg_dump_errors.txt
/pg_dump_errors.txt 文件为两个 pg_dump 命令生成相同的输出，并显示数据库转储，但没有错误，例如
head pg_dump_errors.txt && tail pg_dump_errors.txt
pg_dump: last built-in OID is 16383
pg_dump: reading extensions
pg_dump: identifying extension members
pg_dump: reading schemas
pg_dump: reading user-defined tables
pg_dump: reading user-defined functions
pg_dump: reading user-defined types
pg_dump: reading procedural languages
pg_dump: reading user-defined aggregate functions
pg_dump: reading user-defined operators
pg_dump: creating FK CONSTRAINT "public.timesheet
timesheet_date_linkid_fkey"
pg_dump: creating FK CONSTRAINT "public.timesheet
timesheet_employee_functionid_fkey"
pg_dump: creating FK CONSTRAINT "public.timesheet
timesheet_employeeid_fkey"
pg_dump: creating FK CONSTRAINT "public.work work_categoryid_fkey"
pg_dump: creating FK CONSTRAINT "public.work work_clientid_fkey"
pg_dump: creating FK CONSTRAINT "public.work work_date_linkid_fkey"
pg_dump: creating FK CONSTRAINT "public.work_expenses
work_expenses_date_linkid_fkey"
pg_dump: creating FK CONSTRAINT "public.work_expenses
work_expenses_employeeid_fkey"
pg_dump: creating FK CONSTRAINT "public.work_expenses
work_expenses_expense_categoryid_fkey"
pg_dump: creating FK CONSTRAINT "public.work_expenses
work_expenses_workid_fkey"
我在这里的假设是 anacron（因为 Z-backup 中的 set -e）由于这个原因而未执行，但我不确定，因为我确实已将 stderr 转储到 /dev/null（在 lb 中）。lb 中的所有其他命令都不会生成任何 stderr 输出。
附加信息
-rwxr-xr-x 1 root root 119 Oct 13 13:57 /etc/cron.daily/z-backup
-rwxrwxr-x 1 chh1 chh1 1279 Oct 20 17:26 /home/chh1/bin/lb
-rw------- 1 chh1 chh1 68 Oct 21 06:00 /home/chh1/.pgpass
cat .pgpass
#hostname:port:database:username:password
*:*:crewdb:*:pg#$as(
postgres=# \l List of databases Name | Owner | Encoding | Collate | Ctype | Access privileges
-----------+----------+----------+-------------+-------------+----------------------- crewdb | postgres | UTF8 | en_NZ.UTF-8 | en_NZ.UTF-8 | =Tc/postgres + | | | | | postgres=CTc/postgres+ | | | | | chh1=CTc/postgres postgres | postgres | UTF8 | en_NZ.UTF-8 | en_NZ.UTF-8 | template0 | postgres | UTF8 | en_NZ.UTF-8 | en_NZ.UTF-8 | =c/postgres + | | | | | postgres=CTc/postgres template1 | postgres | UTF8 | en_NZ.UTF-8 | en_NZ.UTF-8 | =c/postgres + | | | | | postgres=CTc/postgres
sudo cat /etc/postgresql/10/main/pg_hba.conf
# Database administrative login by Unix domain socket
local all postgres md5
# TYPE DATABASE USER ADDRESS METHOD
# "local" is for Unix domain socket connections only
local all all peer
# IPv4 local connections:
host all all 127.0.0.1/32 md5
# IPv6 local connections:
host all all ::1/128 md5
# Allow replication connections from localhost, by a user with the
# replication privilege.
local replication all peer
host replication all 127.0.0.1/32 md5
host replication all ::1/128 md5
有人知道为什么这不起作用吗？为什么即使备份似乎进行得很顺利，我仍会从 pg_dump 获得 stderr 输出。非常感谢任何帮助！
答案1
.pgpass是每个用户。运行计划脚本的 Unix 用户与手动测试脚本的用户是同一个用户吗？
为什么即使备份似乎进行得很顺利，我仍会从 pg_dump 获得 stderr 输出
因为-v选项。您应该将其删除，同时删除>/dev/null 2>&1导致可能有用的错误消息消失的，包括在上层调用者中。
答案2
答案是将 pg_hba.conf 更改为：
sudo grep ^[^#] /etc/postgresql/10/main/pg_hba.conf
[sudo] password for chh1:
local all all trust
local all all trust
host all all 127.0.0.1/32 trust
host all all ::1/128 md5
local replication all md5
host replication all 127.0.0.1/32 md5
host replication all ::1/128 md5
现在 anacron 脚本运行正常。