最初我试图将一些项目集合插入到 2 个 redis 集中(也许这根本不是一个好主意,但是......)。我尝试一次添加的条目数量: 750+
现在我在尝试使用 StackExchange.redis客户端执行此操作时确实收到超时异常,有趣的是我能够使用我之前调查过的“旧版”booksleeve 客户端完成类似的操作。
所以,我肯定在某些事情上错了(甚至可能在我最初的 bookSleeve 实现中),只是想弄清楚到底出了什么问题。以下是我与 redis 客户端一起使用的代码示例: BookSleeve:
using (var tran = connection.CreateTransaction())
{
Task lastOpTask = null;
tran.SuspendFlush();
try
{
// perform required configurations/ actions
tran.Sets.Add(_redisConfiguration.DbNumber, CurrentIdsSetDbKey, stringIds);
tran.Sets.Add(_redisConfiguration.DbNumber, CurrentDetailsSetDbKey, stringDetails);
lastOpTask = tran.Execute();
isOperationSuccessful = true;
}
catch (TaskCanceledException ex)
{
...
}
catch (TimeoutException ex1)
{
...
}
finally
{
tran.ResumeFlush();
}
if (lastOpTask != null)
{
connection.Wait(lastOpTask);
...
}
}
StackExchange.redis 实现相同的代码:
var tran = db.CreateTransaction();
// todo: do we need to add here any condition or PipeLining? any watch/unwatch verifications?
tran.SetAddAsync(CurrentIdsSetDbKey, stringIds);
tran.SetAddAsync(CurrentDetailsSetDbKey, stringDetails);
try
{
isOperationSuccessful = tran.Execute();
}
catch (TaskCanceledException ex)
{
...
}
catch (TimeoutException ex1)
{
...
}
开始单元测试后,我收到 StackExchange 客户端的下一个错误:
Message: Timeout performing EXEC, inst: 3, queue: 3, qu=0, qs=3, qc=0, wr=0/0
Source: in StackExchange.Redis.ConnectionMultiplexer.ExecuteSyncImpl[T](Message message, ResultProcessor1 processor, ServerEndPoint server) in c:\TeamCity\buildAgent\work\18a91a3757cef937\StackExchange.Redis\StackExchange\Redis\ConnectionMultiplexer.cs:line 1693
in StackExchange.Redis.RedisBase.ExecuteSync[T](Message message, ResultProcessor1 processor, ServerEndPoint server) in c:\TeamCity\buildAgent\work\18a91a3757cef937\StackExchange.Redis\StackExchange\Redis\RedisBase.cs:line 92
in StackExchange.Redis.RedisTransaction.Execute(CommandFlags flags) in c:\TeamCity\buildAgent\work\18a91a3757cef937\StackExchange.Redis\StackExchange\Redis\RedisTransaction.cs:line 51
in DFS.Cache.CacheManager.RedisStackExchange.RedisContestCacheManager.RegisterAvailableContests(IList1 contests) in d:\Projects\DFS\Code\DFS\DFS.Cache.CacheManager.RedisStackExchange\RedisContestCacheManager.cs:line 131
只是想知道,我到底做错了什么。提前感谢您的任何建议!
PS 对于 stackEchange.redis 配置,我使用的是 Marc 从 github 提供的配置示例(可在此处获得:StackExchange/StackExchange.Redis/blob/master/Docs/Configuration.md)
PS 请查看当前的 StackExchange 客户端配置文件:
var config = new ConfigurationOptions
{
EndPoints =
{
{"MasterIP", 6379},
{"SlaveIP", 6380}
},
CommandMap = CommandMap.Create(new HashSet<string>
{
// EXCLUDE a few commands (to work with data-flow-related mode only)
"INFO",
"CONFIG",
"CLUSTER",
"PING",
"ECHO",
"CLIENT"
}, available: false),
KeepAlive = 60, // 60 sec to ensure connection is alive
ConnectTimeout = 5000, // 5 sec
SyncTimeout = 5000, // 5 sec
ServiceName = "mymaster", // sentinel service name
DefaultVersion = new Version(2, 8, 8),
Password = "password"
};
标准条目(用于集合)如下所示:
{
"Id":"08e5ffdbced046cb8f55c50e4bab822d",
"Entries":0,
"State":"i",
"Name":"very dummy entry name value: autoGet 299",
"Summary": "entry summary details, some long string 299, some common info, some data: true, 8200"
"IsMultiple":true,
"IsPublic":true,
"MaxEntries":8200,
"IsEntryVisible":true,
"StartDate":"9/10/2014 12:00:00 AM"
}
PS 在 Marc 的回复之后,我运行了几个测试并在单元测试中遇到了一些其他错误,例如
{"Timeout performing SISMEMBER set:raw:Ids, inst: 1, queue: 6, qu=0, qs=6, qc=0, wr=0/0"} when set contains not more than 300 items.
所以我同意这是连接问题,与当前切换到 StackExchange.redis 客户端无关。